QuarkChain FAQ

quarkchain (40)in #ama • 6 years ago (edited)

Part 1: Marketing Questions

Q: There are so many blockchains these days and they are quite competitive. What plans does QuarkChain have in place to encourage the community to support this project continuously?
A: We will continue to post our development process, ecosystem building and many more on our social media including Twitter, Telegram, Medium, Steemit, and Reddit. Except for previous 100+ volunteers helping us test our testnet, since our testnet 1.0 has been released, there are more than 3000 community members have joined the testing. We also have developer communities which are under development.
Q: Can you introduce your partners?
A: We have built strategic partnerships with 30+ global projects such as Tripio, Bodhi, and Laya.one. We also have plans to build deeper relationships with 10 projects including Covalent Chain, DxChain, Drep, Playtable, ValPromise, Ankr, MXC, LendChain, EON, and Celer. Besides, we also partner with Certik in Smart Contract audit. More partnership will be built.
Q: What’s next in the roadmap?
A: We will introduce our next plans in three major parts.
1. Development
  The first thing we need to do is to make sure our testnet is stable and keep optimizing our systems. We have found that there are many places, not only in scalability part but also in virtual machine and storage part, that we can improve in the following several months. We are also preparing articles of our technical details for open source several months later. We want to encourage community members to participate in our project and make our project not only our own project but also the community’s project. Another big thing we are focusing on currently is our mainnet which will be launched in several months. The main feature of the mainnet is that we can increase capacity on-demand as the network grows, and it will work as a scalable smart contract that can do whatever ETH can do but with greater scalability.
2. Marketing
  Currently, we only separate our market into Chinese, English, Korean, Japanese, Russian parts. We will have more strategies to open for different markets including, Thailand, Vietnam, Singapore, India and Europe. We will do more local stuff and enlarge our local community. Moreover, with the launch of testnet, we will build developer communities. At the beginning of August, we are going to hold the biggest hackathon in the Bay Area with Google ABC. There are only three projects to be selected and we are very honored to be one of them. At that time, there will be many programmers from big companies such as Google, Facebook and Linkedin building dApps on top of us on this two-day hackathon. We also have our 50 million eco-fund to establish an open and collaborative ecosystem of QuarkChain and 30 partners after just one month on Binance.
3. Korean Marketing
  We recently had the signing ceremony with a very strong insurance company in Korea who has revenue of 20 million per year and decides to go blockchain and global. We also have several contracts ready including a leading AI company and leading financial institution in Korea. You will hear more news about Korean marketing very soon.
Q: Why the current circulating supply seems too low compared to the declared total 10 billion circulating supply?
Please note that 40% QKC will be used for MINING and is already locked by Smart Contract. Private sale is locked to protect public sale investors. The first release of private sale is 10% and it will be released in about ONE MONTH after the QKC is listed on exchange. You could see the circulating supply schedule detail here:
https://support.binance.com/hc/en-us/articles/360004471832-Binance-To-Open-Trading-For-QuarkChain-QKC-and-Risk-Warning
Other token allocation includes 15% for the team, 15% for the foundation, and 5% for advisors. These are all locked up to 2 years with vesting plan using smart contract and will be unlocked gradually.

Part 2: Technical Questions

Q: What kind of language is QuarkChain using for development?
A: Currently, QuarkChain is developed in Python. The main reason for choosing Python is its fast deployment so that QuarkChain team could focus more on technology. Actually, we already obtain pretty decent performance results these days, and we could easily achieve much higher performance by employing other high-performance languages such as C++ and Go. Note that early Ethereum development also used Python, but later Go implementation becomes popular after Ethereum got more attention.
Q: What does Collaborative Mining of QKC means?
A: QuarkChain will utilize GPU-friendly mining algorithms, which is still under development. QuarkChain Network has several minor blockchains (shards) and one root blockchain. Each minor blockchain offers different incentives and difficulties. Miners could choose any minor blockchain at an optimal price of their hash power. This creates an open market economic model, where a blockchain is a seller with goods being the block reward, while a miner is a buyer with hash power being their currency. It is desirable that a marketing model is designed with features ensuring that though each party in the market pursues their interests, the collective behaviors of each party can benefit all. The goal of collaborative mining is to design incentive mechanisms and difficulty algorithms so that
(1) Hash powers are incentivized to distribute evenly among shards. This ensures that all shards are mined evenly, and thus the system throughput (i.e., TPS) increases as the number of shards increases;
(2) The root chain has a significantly large portion (over 50%) of hash power over the whole hash power of the network. This prevents double-spend attacks, and a malicious miner needs at least 50% * 50% = 25% power to perform an attack.
Q: What is QuarkChain’s relationship with DAG or other Tangle technology?
A: “The tangle is what is known as a directed acyclic graph (DAG): a data structure that moves in one direction without looping back onto itself. ” (from https://www.nasdaq.com/article/what-is-the-tangle-and-is-it-blockchains-next-evolutionary-step-cm911074)
The system of QuarkChain Network itself can be treated as a well-structured DAG. This allows QuarkChain to inherit a lot of benefits from both blockchain and general DAG technique. For example, the consensus of QuarkChain and its threat model can be easily derived/analyzed following those of Bitcoin/Ethereum blockchain, while QuarkChain achieves high throughput similar to general DAG. Given two blockchains/DAGs of QuarkChain, we could easily tell which one should be appended thanks to QuarkChain’s root chain.
Q: How does cross-shard communication work in QuarkChain?
A: The QuarkChain Network fully supports cross-shard transactions as the first-class citizen, in a sense that:
(1) Any user could issue any cross-shard transaction at any time;
(2) Cross-shard transactions can be confirmed in minutes;
(3) The throughput of cross-shard transactions could be scaled linearly as the number of shards increases.
In short, the cross-shard transaction is almost the same as in-shard transaction except that the root chain needs to confirm the block header of the transaction before spending the output of the cross-shard transaction.
Q: It seems there would be different nodes with different roles, all interconnected. How do you plan to prevent them from exploiting the role-playing model? As I understand it, you will manage and audit the network of voluntary nodes, then how do you call it “public blockchain”? Also, sharding doesn’t guarantee the persistence of data, nor completeness of the collection of shards. How do you guarantee longtail operation will be smooth and stable? What if there aren’t enough volunteers to participate?
A: (1) For the first two questions, nodes (machines) trust each other to form a cluster acting as a full node. Anyone can run their cluster to participate in the network. Thus, we don’t manage clusters directly;
(2) For the third question, there will be data completeness for an individual shard. Sharding and persistence are not mutually exclusive and we don’t understand why you think sharding doesn’t guarantee the persistence of data. All major data stored in Amazon, Facebook and Google use sharding to achieve scalability, and we are pretty sure persistence is guaranteed;
(3) For the last question, mining is about incentives. We can try to solve the cold start problem by encouraging mining with relative greater incentives at the beginning.
Q: Is that possible to say a dApp to seamlessly run on multiple shards if one shard cannot provide the necessary throughput? If that possible, as cross-shard transactions are slower, wouldn’t that create somewhat of a bottleneck as well?
A: There is a topic of a scalable smart contract. We are working toward this feature, and a lot of interesting things are ongoing. Also, it depends on how the dApp is configured as well. Take CPU as an example, once Intel/AMD reached the clock speed limit, they realized multi-core should be the next design paradigm, which means performance software should also change the paradigm to fully leverage multi-core CPU architecture.
Q: Number of Nodes - Can you explain to me if the more nodes, the better? Is that possible for QuarkChain to reach high TPS with fewer nodes (to prevent slower network)?
A: It depends on how these nodes are organized. If all nodes would like to reach the same chain consensus, then the more nodes in the network, the slower the network is. Generally speaking, the more nodes in the network, the more decentralized the network is. Thus, we could achieve the high TPS with fewer nodes, but this will sacrifice decentralization, which is what we want to encourage. This shows the trade-off.
Q: Number of shards - How does the number of shards are selected, how many nodes will be there in the number of shards? As per the white paper, each shard will have its difficulty and reward mechanism. How is it defined? So it means miners can switch over between the different shards depending on mining difficulty and can try to get maximum rewards? How is this mitigated? Is there any sort of EDA or there is a limitation for miners switching between shards? How is this more decentralized than usual PoW solution?
A: The number of shards is determined by the network situation and could be done by our governance model. The miner could mine any shards, depending on block reward, difficulty, and network propagation of the major miners of the shard. More decentralized is mainly because a miner could mine a shard directly instead of joining a mining pool. The motivation for joining a mining pool is to collect reward timely as an exchange of transaction fee of pooling. By mining the shard directly (as the difficulty is lower), the miner could save transaction fee and encourage more decentralization.
Q: Clustering - It is a good idea where the "honest nodes" are clustered to run as a supernode and will involve the root chain to confirm the transactions between them. There will be the incentive for the nodes to form clustering. How does this "Honest nodes" are selected for clustering or is it something which the nodes can do themselves? If they can do themselves? What prevents the malicious miners to collude and form a cluster of their own? How is this mitigated?
A: A cluster is a replacement of a super-full node, but still serving as a peer in the network. Therefore, as long as there are sufficient peers (clusters) in the network, any blocks from the malicious cluster (peer) will be rejected. At the moment, a smart contract can be only administered in one shard. A cross-shard transaction is to transfer QKC from one shard to another shard, and thus a user with a single private key will be able to execute a smart contract transaction in any shard. A cluster - as a replacement of a super-full node - maintains the full ledger of the network and thus knows all chains. In addition, double spending attack is mitigated by root chain's hash power via root-chain first consensus algorithm. Please refer our white paper for more details.
Q: Does QuarkChain have any plans to move away from the EVM for dApps with many other VM's coming out, such as NEO's VM. Or do you intend to create your own VM?
A: We may develop our own VM if needed, but this highly depends on the feedback of our dApps partners. Even though there are so many VMs, a lot of them lack systematic supports (such as editor, compiler, debugger). To our best knowledge, EVM is the most-adopted VM right now, and other candidates could be NEO VM, EOS VM, and ETH WASM. Currently, we don’t have the plan to swap VM but will add more supports for new VMs, i.e., adding new shards to support new VMs or even new consensus algorithms. This shows another advantage of our sharding technique on enabling this flexibility. In this situation, QKC will be the GAS, and other VMs may have different token models. We need to figure out the proper way to incorporate them. However, this should happen after the launch of mainnet.