JustPaste.it

NKN AMA questions & answers

1) Briefly tell us about yourself

Hi, I’m Yanbo, founder of NKN project. I love coding, and have done quite a few open source
projects both in blockchain as well as wireless networking stacks. In the last two years, I’m the
co-founder of OnChain and core developer for DNA (Distributed Network Architecture) open
source project. I lead a team of highly talented developers in Beijing and California. As a hobby,
I practice boxing which keeps my mind and body sharp! So in a sense I’m more Rambo than
Lambo.

2) You are the founder of NKN. How do you came up with this idea?

In the past few years, we have been engaged in the development of blockchains and have
completed quite a few blockchain systems. However as I mentioned earlier, we have been
doing decentralized peer to peer network research and development long before doing
blockchain. For example: WiFi-based mesh networks, and some actual decentralized data relay
and transmission system deployment.

Because one of the core principles of the blockchain itself is to decentralize the network, we
have been thinking about how to use this feature of the blockchain better to solve the problem of
idle network connectivity and transmission capability. Through blockchain's self-incentive
mechanism, there is an opportunity to promote the continuous development and expansion of
the decentralized network. At the same time, we also see that there is currently no blockchain
system that can do a good job of implementing network resources tokenization.

Well, we are wondering, can we combine this with our background and expertise to make a
network system that everyone can participate in and build it together? That will fundamentally
solve many of the problems of today’s Internet, such as network fragmentation and network
neutrality. So, here comes the NKN project.

3)* Does Onchain support you desires?

When I worked at OnChain, I had great time leading a team of talented engineers to develop the
open source DNA (Distributed Network Architecture) software. It not only gave me technical
experience in Blockchain but also taught me how to build ecosystems and how to work with the
community.

However, OnChain focuses on blockchain solutions for large enterprises with a different design
philosophy than a fully decentralized public blockchain. Therefore, NKN is a completely
independent project, operated and managed independently of OnChain and NEO. For example,
this is totally different from the Ontology project.

We are excited about the great interest and early support of Mr. Da Hongfei, founder of NEO.
And we are very pleased to enlist NEO Global Capital as one of the key strategic investors.

4) You need a pretty big team to realize your idea. Not only architects and seniors
presented on website, but developers as well. How big the team is? Is it centralized
geographically?

You are right that this is an ambitious project and needs lots of talents. Our solution is not an
army of software engineers, but rather a combination of excellent core developers, open source,
re-use of existing solutions where applicable, and a network of community contributors.

We have about 10 developers already, who have extensive coding experience in blockchain
and networking software stack. Our team are distributed between China and US. In addition, we
have researchers in universities and research labs such as Dr. Diffie’s Cryptic Labs that are
working on the fundamental cryptography and algorithms. These are the key talents form our
core developer group.

We will not reinvent the wheels, and will use open source components and existing algorithms
in conjunction with our own contribution. We believe in building on the shoulder of giants. In
addition, our previous code base in DNA can be used to speed up development in the initial
prototype and Proof of Concept phase. Although long term, many of the core components will
be rewritten from ground up. But it gave us a sandbox to do key new component development.

We will also strive to build a vibrant open source community, where community developers can
contribute to the NKN open source project where they are most interested. This includes not
only implementation itself, but also the detailed design. And they can be rewards either in
tokens or kudos.

5) What issues did you struggle with in the current phase of development/designing?
What seemed to be the most difficult one?

Intuitively, it is very natural to combine the network resource sharing with blockchain’s incentive
mechanism. And based on our previous experience in both fields, we believe we can make
them work very well together.

However, the biggest challenge is that some malicious nodes will create fake traffic and profit
from it. On the consensus algorithm, we also need to take care of all the potential attacks and
corner error cases that are either unintentional or malicious. I think we are spending probably
more than 90% of our energy on those challenges such as fake traffic and attacks.

6) You’ve mentioned ‘cellular automata’ in your whitepaper. Could you tell us more?

Cellular Automata is a concept we have been very interested in for a while now. In layman’s
terms, it is a powerful tool to model complex systems by using simple rules. It has been applied
in artificial intelligence such as neural network, network routing, and cryptography.

Cellular Automata has a few unique characteristics that are extremely attractive for network
routing as well as scalable consensus algorithm:
- Truly equal peers
- Simple rules
- Native parallel processing
- Dynamic
- Decentralized: local decision, global impact
- Open and expandable

By finding the right rules and initial state, you can classify the system into 4 kinds of states.
Class 1 “Homogeneous and Stable”, is very useful for consensus. Class 4 “Edge of Chaos”, can
be explored for dynamic routing with some randomness.

We have witnessed failures of many over-complex software systems. Cellular Automata inspires
us to solve complex network routing and consensus problem not by modeling the entire system,
but rather focusing on the simple rules that define how individual nodes interact with each other.
This is the essence of decentralized system, and Cellular Automata concept fits right in.

6.1 On what phase of development is it (cellular automata)?

Cellular Automata is a concept and methodology that will guide the design and implementation
of NKN protocol and system. So it will be useful for dynamic network routing as well as
consensus algorithm.

Our additional contribution is using the Ising model in Physics to mathematically prove that
whether a CA rule is convergent, and also mathematically guide us to find the suitable rules.
Thus we do not have to exhaustively searching among millions of potential rules.

6.2* What upsides does this technology have? (the answer has to be about channel
capacity)

The model based on Cellular Automata has some outstanding features, such as true
decentralization and equality between nodes, which can run concurrently and is very easy to
scale. These features are very much in line with the NKN network we are building. At the same
time, the rules of interaction between individuals are relatively simple and easy to define, which
contributes to straightforward and robust implementation.

Cellular Automata has a few unique characteristics that are extremely attractive for network
routing as well as scalable consensus algorithm:
- Truly equal peers: decentralized, no dependency on seed or core nodes
- Simple rules: easy to implement, simpler code, less errors, consume less resource so
potentially working on billions of mobile devices
- Native parallel processing: highly scalable
- Dynamic: network topology can evolve with a touch of randomness, difficult to detect
network traffic pattern, more resistant to attacks
- Decentralized: local decision, global impact
- Open and expandable

7. How dApps would interact with your layer?

dApp can use NKN API to utilize our network connectivity and traffic relay function on the
networking layer. In addition, dApp will interact with the incentive layer for billing or mining. So
dApp can be both consumer of as well as supplier for NKN network. In the future, we can
explore providing NKN protocol to other blockchain system as their network layer. We can also
explore opportunities to use NKN as a platform to dApp to issue their own token and write their
own smart contracts. But those are longer term possibilities, and not our current focus. We need
to make the foundation very solid.

One concrete example of what dApp can benefit from NKN is truly P2P messaging application.
Today, because most of us are behind NAT (Network Address Translation) with private IP
address as well as firewall, we have to rely on a central server to relay message. This applies to
WhatsApp, FB Messenger, and Telegram. By using the NKN network, a messaging dApp can
enable decentralized P2P messaging from day one without its own server infrastructure.

8. Could you tell us about the mining economy of your network and what role do tokens
have in it?

The way NKN network incentivize nodes to relay traffic is similar to what IPFS/Filecoin did to
incentivize decentralized storage. There are two ways the node gets token rewards: one is by
providing connectivity and relaying traffic for other nodes, another is by mining, i.e. helping
record transactions on the block. The amount of tokens a node can earn is proportional to the
amount and frequency of their contribution, with some probabilistic algorithms. We are
designing an economic model that will encourage more nodes to join the NKN network, share
extra network capacity, upgrade transmission throughput and reduce latency, and expand the
network. And due to the incentive model, NKN network will be more stable and attack resistant
once more nodes join the network.

9. Are you planning to integrate into existing platforms (NEO/ETH) and how?

NKN is an independent project that need to revamp the plumbing of both network layer and
consensus layer. Therefore it is not built on top of NEO, ETH or other existing public blockchain.
NKN will have its own public chain to record and clear transactions for network connectivity and
data transmission. There are a few reasons for that:

1) Efficiency of verifying transaction: a data communication network on a global scale will
generate large amount of transactions in real time, whether it is session based or packet
based. We have not found an existing blockchain system to handle that.
2) Number of nodes to reach consensus: we are talking about at least millions and more
likely billions of nodes reaching consensus. In contrast, Bitcoin and Ethereum have a few
thousands nodes. We need to do some hard work at the foundation of blockchain
consensus algorithm to make it work. Cellular Automata is one of the concept we are
leveraging.

Overall, we want to have more nodes to join the NKN network, both for increasing network
capacity, as well as improve stability and attack resistance. And the tokenization of transmission
capacity will incentivize such behavior. So the dynamic network routing and underlying
blockchain consensus algorithm has a symbiotic relationship. Both need to be re-architected,
redesigned and rewritten to optimize the sum of two.

10. What are your fundamental goals?

Our slogan sums up our fundamental goal quite well “Free the bits, rebuild the Internet we
always wanted”. To break down into more specific goals:

We want to
- Connect anyone at anyplace
- Secure net neutrality from network layer
- Make the Internet always open and scalable
- Reward resource sharing and improve connectivity for everyone
- Build the world’s largest Internet bandwidth exchange

11. What phase is your project on?

We are approaching the end of phase 2 in our roadmap. We published our whitepaper last
month, which defines our overall direction, main technical concepts and high level designs for
implementation. Right now, our focus is on low level design (LLD) and coding implementation in
parallel. We are building a prototype system, leveraging our previous work in DNA as the
scaffolding framework. We are tackling the most challenging topics in routing protocol, Proof of
Relay, and consensus algorithms. Both the LLD and source code will be published on github
under the main branch in coming months. So community members are welcome to review and
contribute to the project. We want to be fully open and transparent.

On the business side, our economic model and token metrics have been shared with our
earlybird telegram group. We look forward to sharing with all of you real soon now.

11.1 Do you have results of some alpha-tests that you want to share us?

We are not ready for alpha-test yet. Our current roadmap estimates that we will have testnet up
and running during Q1’2019. As we mentioned in earlier questions, we are currently developing
our prototype system and Low Level Design docs. Hopefully we can share these intermediate
results with the community early on, even though these might not be the final design and do not
have all the features.

12. What is your key objective?

See our reply to question 10.

13. In your whitepaper you write that information can be censored and privacy protection
is limited. This problem is obvious consequence of network centralization, in its turn,
centralization occurs because of the physical (geographical) location of the nodes and
the cables that connect them (I am connected to a third level provider, this provider to a
second level etc...) and its impossible to change. Under these conditions is adjusted the
last level (layer) of TCP/IP model (physical), but in wp you write that you realize network
model from application to network layers, hence the question is how do you want to
solve censorship problem and similar which in the first place are the consequence of
physical network?

We understand that we cannot change the physical networking infrastructure overnight.
Therefore in short to medium term, we are working on a software overlay approach which is at
IP routing layer. This pure software approach will be compatible and adaptable to all kind of
underlying physical infrastructure, and speed up the initial adoption and ramp up of NKN
system. In this phase, we can still establish a more dynamic and more random logical network
topology over the same physical infrastructure. And help users to enjoy safer and more ways to
reach all kind of servers and services.

In the long run, we do intend to use the incentive mechanism (both Roof of Relay as well as
block generation) to stimulate new hardware and new infrastructure build-out by the miners and
users. Then we will be beyond overlay network, and truly revolutionize the network
infrastructure. We hope it will create not only new suppliers , but also richer and more dynamic
topology.

Elon Musk’s SpaceX is building a new satellite communication network, imagine what can be
achieved if his satellite links can run NKN protocol and become part of NKN network? When
new hardware, new software and new incentive model work together, we will see a very exciting
the different future.

12.1 Does Node of your network have to be high productive? Or is it possible to use
‘weak’ devices like routers and modems as a Node (firmware is implied)?

No, the NKN software will not require high processing power and can be run on personal
computers, smartphones, routers and switches. That is one of the design goals of NKN, and the
reason we are putting serious research hours into Cellular Automata. Naturally the more
processing power and link capacity a node has, the more reward it can generate from relaying
traffic and help record transactions.

We might be able to run the software on even smaller embedded devices (e.g. IoT devices), as
long as they run on Linux. Our initial software overlay should be able to run on any Linux system
in user space. We do not intend to build the software in Linux kernel space in the very
beginning. Even though that means greatest efficiency and potentially smallest footprint, it will
complicate the deployment model and slow down initial adoption.

In the long run, we do intend to have deeper stack integration and optimization with network
equipment vendors, down to hardware and even ASIC level. That will give us the quantum jump
into next level of transmission capacity and user experience for the NKN network.

14. How big is the industry that you want to enter(network industry)?

According to Gartner’s report, global communications/network services will be about $1.4
trillion-dollar annual business in 2018. And this is an industry that has not been disrupted for
over 50 years, therefore a huge potential when NKN innovate with both blockchain technology
and tokenized economic model.

15. What are the benefits that are not found in your competitors?

People often ask us, what is different about NKN compared to other projects in the networking
space. Some projects are entirely focusing on mobile phones and using Wi-Fi, Bluetooth, and
Wi-Fi Direct. NKN intend to change the entire Internet, including fixed and mobile, phones,
laptops, edge and core network. NKN is a protocol overlay that is independent of the underlying
physical communication media, and transparent to applications.

In addition, most networking projects are based on Ethereum, ERC-20 tokens and smart
contracts. We are not sure if Ethereum will be scalable for millions or billions of nodes, and
petabytes of traffic. NKN will have its own blockchain technology that will have fundamentally
new Proof of Relay and new consensus algorithms that will be newly developed specifically for
tokenizing the network transmission capabilities.

Compared to some projects, NKN have a much broader scope. We want to tokenize networking
like what IPFS did to storage. Our target is the 1 trillion dollar communications business, and
how we can provide better connectivity to everyone by incentivize the sharing of unused
networking resources, expanding and upgrading the sharing network.

16. Some projects are immersing themselves into development and forget about the
marketing. How important is the marketing in your project?

A project will succeed only if it provides value to average users and have economic benefit for
all participants. Just cool technology is not enough. Therefore we have a solid business and
ecosystem development plan, created by seasoned industry veterans. I believe that’s what you
refer to marketing? We are definitely not hyping our project.

To build the NKN ecosystem, we need to work with Device and Infrastructure equipment
vendors, Internet Service Providers, and dApp developers. With the right economic model,
these four elements will combine to boost the rapid expansion of NKN network, faster innovation
cycles, and better infrastructure and services for everyone.

16.1* Do you aim to form a big community?

Absolutely, we would love to form a big community. We are not focusing only on making the
community bigger, but more importantly a higher quality community. That community will
embrace technology, use, improve, and expand NKN network, participate in NKN core software
and application development, support new users.

17. What hard cap is in your project?

10% of entire tokens are offered to earlybird/pre-sale and crowdsale, while another 25% for
private investors. We are pegged to ETH instead of fiat since the exchange rate has been very
volatile. You can find all the token metrics in our official Telegram group.

17.1 Do you familiar with any metrics on private? (bonuses, token lock)

We typically cannot disclose the details of private sales, due to NDA. But it suffices to say that
we do not need to give very steep discounts even for large strategic investors. The reason is
that we are entirely self-funded during seed round. We are well known in crypto circle: major
investors in this space knows our personality as well as our work.

In addition, we typically have 6+ months vesting period for private investors. In some cases,
almost as long as The team/developer.

You can find all the token metrics in our Telegram official group.

17.2 Is there the token lock for a team? If yes, how much?

18% of NNC will be reserved for developers and advisors, and will be transferred to them over a
3-year period. Upon the expiry of 1 year, the team will receive 1/3 of this allocation, with a
further 1/3 of this allocation to be distributed upon the expiry of each year thereafter, until the
3-year period is up.

We have set up 3 year vesting period because we have confidence in our project. In addition, as
an engineering team we would like to make sure our system is reliable and dependable for long
term use.

17.3 The trend is to sale tokens straight off on stock exchange to avoid legal issues. Will
there be the token sale?

We are working closely with our legal team to ensure full compliance in major regulatory
regimes. So far we have carried out our private sales towards accredited institutional investors
directly, while delegate earlybird presale and crowdsale to LeekICO private placement. For
instance, we do not allow US, China, Singapore, and Australian citizens to participate.

18. How much time is needed for final implementation?

According to our roadmap, we plan to release the TestNet during Q1’2019. After enough time to
validate and improve the TestNet, MainNet is targeted to go live during Q3’2019.

During 2018, we will open source both the detailed design as well as the early prototype of NKN
core protocols. We welcome community contribution!

Additional questions

A1. What is the network launching process and what are the main issues, that can be
encountered?

Like most open system, it will be under all kinds of malicious attack as soon as it is online.
Because the economic incentive for doing bad things is high. Therefore we will engage
community from very early on, to participate in our detailed technical design and prototyping
during the rest of 2018. This way, we can try to address most known attacks in advance.

Then once we launch the TestNet during 1Q’2019, we will start with certain entry barriers and
limitations to gradually adding NKN nodes in a controlled way. Later, we can gradually move to
fully open MainNet once we have enough “good” nodes to counter most attacks.

A2, Is Whitfield Diffie just an adviser? Did he participate in the research part of the
project? If not, how does he evaluate your conception and adopted models?

Cryptic Labs and its Chief Scientist, Dr. Whitefield “Whit” Diffie recently confirmed that they have
signed research agreement with NKN to accelerate innovation in cryptography research. As part
of the agreement, NKN is honored to enlist Dr. Diffie as its technical advisor.

As part of the agreement, Dr. Diffie will oversee the research effort by the joint team between
NKN Labs and Cryptic Labs. There will be regular research review meetings, as well as ongoing
email discussions. The researchers in Cryptic Labs, lead by Dr. Diffie, will carry out NKN’s 2018
Research Plan. For example, one of the first research topics that Dr. Diffie and his team are
working on is a better signature chain for the Proof of Relay work.

A3. Your project’s aim is to make internet decentralized, that contradicts the interests of
government regulatory agencies. What are the issues to be encountered? How to deal
with them?

I think this depends on the economic model and how to get win-win between different
stakeholders. FCC in the US, for example, historically has been a strong supporter of network
neutrality, unlicensed WiFi, and more competition in the communication service business.
Sometimes these regulatory agencies want to stimulate innovation and thus economic growth in
their own country, If blockchain and decentralization is the future engine of growth, I trust some
of the wiser government agencies will embrace projects like NKN.

On the other hand, market will prove if a new technological revolution is successful or not. The
typical example is that social APP like WhatsApp and WeChat are replacing the traditional
phone call and SMS market. Wise regulations still play a critical role in stimulating the innovation
economy.