social social social

Matic Network’s first Technical AMA summary — Delroy Bosco

Polygon Team
Polygon Team
Aug 18, 2018
Matic Network

Dear Community,

Thanks all for participating in our Telegram group’s first AMA. We really appreciate all your support. There were really good and thoughtful questions that were asked. For those who missed our AMA, this blog post is for you ?

AMA Starts now!Q. Can you tell me more about how you’re using Plasma to scale?

A. We are implementing the Matic Network infrastructure on the foundation of Plasma. The main way to incentivize that Plasma operators (in our case, this is a 2-layer Delegate and Staker chain for added security and mitigation of block withholding) remain non-Byzantine in block generation is an ever-running bounty campaign on the Plasma contract in the Ethereum mainchain.

For this, Plasma framework mandates proofs of the transactions/blocks produced on the side chains to be published on the mainchain. The public checkpointing layer will validate all the transactions happening on the side chains and publish the proofs to the mainchain. To ensure foolproof security of side chain transactions, the mainchain Plasma contract contains various kinds of Fraud Proofs where any sidechain transactions can be challenged for any fraudulent activity. If challenger succeeds, the stakes of the side chain actors involved in the fraud are slashed and are provided to the challenger.

Q. How Matic is going to protect double spend while exiting on mainchain?

A. So Basically, every time a user wants to withdraw, he has to initiate the withdrawal on the mainchain. Once the withdraw has been initiated, there is a waiting period of 7 days, and exits are based on a priority queue. During this withdraw period, anybody can come and check that if there’s a fraudulent activity on a particular exit, and they can challenge it. If the challenge is successful the faulty withdraw claim is discarded. Hence double spends are mitigated. For better user experience we are working on fast exiting mechanisms which users can use to avoid the wait periods for withdrawals.

Q. What’s the Delegate layer?

A. At the Matic blockchain layer, we have Delegates selected by PoS Stakers on the base layer with Proof of Solvency who will be creating the Matic Blocks. To achieve faster block generation times, these Delegates will be low in number. This layer will achieve ~1 second block generation times at extremely low to negligible transaction fees.

Q. Please tell us about your competitors — Celer, Loom, etc, strong and weak sides?

A. Both Matic Network and Celer Network are different solutions to the same problem — low transaction throughput in current blockchains. Both utilise off-chain scaling techniques and rely on the main chain for final security; however the fundamental difference is in the approaches — Matic Network is based on a set of Plasma sidechain(s) backed by Proof-of-Stake consensus (see for more details), whereas Celer Network is a state-channel based solution. Both projects aim for generalized state transitions off-chain but in vastly different ways.

Matic Network is aiming to build a dApp developer ecosystem. Since it uses an account-based Plasma sidechain and also employs an EVM-compatible runtime known as the Matic VM, it will be relatively easier for Ethereum based dApps to migrate to Matic Network once it is live. So in this respect as well, Celer Network is different in terms of developer interfacing. There are other differences as well — let’s see if we can publish these in a long form blog post.

For Loom, I think first and foremost, their focus is different. Loom is focusing on games and social apps with less decentralization while Matic needs security and decentralization as we are gearing for both normal DApps, games, financial transactions and trades. We also have plans for full-blown financial services like lending/trading DApps (token-sets swaps, margin trades and much more)

Secondly, Plasma Cash, which Loom wants to use “in future” block times will always be more than the Ethereum block times as you need to push every block of the sidechain to the main chain, while Matic uses checkpoints(with PoS layer) for 1-second block times (with DPoS layer)

As Plasma Cash works with the NFT (Non-Fungible Token), it works great for game cards and social state changes where you have pre-defined fees (bundled as NFT — eg “20 tokens” to play game == 1 NFT coin on plasma cash). For normal tokens transfer, you may need to swap tokens (like normal currency notes change) on top of Plasma cash which I think makes it difficult (more difficult for usability/users). It is still being discussed on Plasma calls. While Matic uses state-based plasma (closer to Plasma MVP), this is not the problem for us.

Q. I really like the concept of ethdagger, Can you tell us more about it and the other products under development?

A. Dagger is part of our initiative to help developers create great dApps. It is an easy way to get realtime events from the Ethereum blockchain. As you know, the development tools on Ethereum and other blockchains are still getting developed and maturing day by day. So it is our contribution to the community. We are also developing several new tools for better developer experience.

We are also developing a wallet implementing the WalletConnect protocol, which is an open protocol to connect web-based distributed applications to mobile crypto assets.

This wallet will help users to interact with DApps and sign transactions easily, while still keeping their private keys safe on their mobile. This should go a long way in making blockchains accessible to mainstream users.

Other than this, we are also looking at Context specific ether fewer accounts and Gas relay abstraction on identity to enable ether-less sign transactions, which can be a huge booster for mainstream user adoption.

Q. There is an opinion that the Ethereum sidechains don’t need their own token.

A. This is a very good question actually. There are following key considerations :

1. We intend to build Matic as a general purpose scaling solution for public blockchains. As in future, there can be multiple base chains like Ethereum, its not favourable to build the Staker ecosystem on any other asset than Matic itself.

2. Appcoin security model — i.e when the consensus participants have their stakes in Matic tokens, they have a strong economic disincentive to not act in the disinterest of the network

3. No dependency on Ethereum as if in future there’s some existential threat on Ethereum, Matic token holders and stakers can simply move to other blockchains like NEO and EOS. While if we allow people to stake only in Ether, then it makes us platform dependent

Also, a practical issue is also that since the PoS ecosystem requires high value stakes to safeguard the network, Staking in Ethereum or any other base chain would make it financially not viable for us to setup the base blockchain infrastructure.

Q. “Once deposited tokens get confirmed on the main chain, the corresponding tokens will get reflected on the Matic chain.” if we deposit ETH on main chain contract will the ETH will be reflecting on Matic? and what will be the use of Matic chain token?

A. In answer to your first question, ETH deposited on main chain contract will be reflected as WETH (wrapped ether) ERC20 tokens on Matic. All tokens on mainchain are represented as ERC20 tokens on the Matic chain.

The second question has been answered above.

Q. Celer has a lot of public strategic investors already before ICO. Who is supporting matic? And more importantly, did Matic Network have a pre ICO sale?

A. Celer started their ICO journey quite long back than Matic Network. We will be getting there, and getting there real good :). Just bear with us, these things take time in terms of their finalizations.

Q. Will there be KYC before selecting the delegates?

A. Our core objective is to have pure Economic incentive/game theory model for the network to achieve its strength in terms of security and self sustenance. KYC and other methods are just add-ons which we will use if we want to have an additional extra-network layer for further cushioning of security. The current focus is to device autonomous economic model for the consensus participants to join and participate in network progression.

Q. “EVM-compatible runtime known as the Matic VM”, it will be relatively easier for Ethereum based dApps to migrate to Matic Network once it is live. So how you will differentiate Matic as general purpose scaling solution? how will Matic VM (Ethereum VM compatible) work with NEO or EOS?

A. The initial objective is to achieve scaling for dApps on the Ethereum platform, with a short to medium term focus on enabling assets interoperability between other chains such as NEO, EOS, etc. We will get down to solving for general cross-chain smart contract interoperability later — i.e. a general smart contract execution engine. We want to ensure we prioritize our roadmap appropriately and you will see a slew of initiatives on this in the future from us.

Q. Celer has nicely packaged old tech with cChannel, cOS + too much marketing source and there is nothing new and progressive…What do you think about Celer’s technology?

A. Haha, this is a public forum. Lets not take direct jibes on other projects ? ?

But having said that, State Channels, specially Generalized state channels have been an area of interest for a long time in Blockchain community. Many projects have attempted it but haven’t been successful yet. Wishing all the best to Celer to solve it. Its a big problem, if anyone is able to solve it sustainable and with a good Developer and User experience, it will take the entire blockchain revolution to next stage

Q. It just adds confidence to the community if some big names back the project. Can you tell us more about the economic incentives of Matic, As in how will the token economics work?

A. Definitely. Backing from big funds adds a confidence. And we should also be able to garner the similar, if not better interest from the community. Though market conditions are way different than what they were 1 month back, good projects will get ample backing from the market.

In terms of economic incentives, The block producers in the sidechains get rewards in terms of the transaction fees. The second layer of checkpointers/validating nodes also gets a cut for putting the checkpoints (proofs) after going through the consensus (on this decentralized layer of checkpointers).

Q. Can you explain the impact of Casper on sidechain projects which may go live by late 2019–2020? Also, how does Matic differentiate themselves from Loom network

A. Casper, Sharding, and Plasma are the 3 main initiatives from Ethereum for their near-to-medium term roadmap. EF has specifically mentioned that they will not develop Plasma on their own, and they want the community to build solutions for it. They are pretty clear on that.

Casper is a consensus mechanism change and will have little impact on scaling. Sharding is definitely very interesting. But even with sharding, the capacity is not going to be huge — ~1500 tps or so. Even a decent sized application will get choked on this.

To better understand the scaling context, the base chain need not necessarily be ultra-fast — but it does not need to be extremely secure and decentralized. The sidechain architecture, especially Plasma, is the correct way to go ahead with scaling because you need an absolutely strong Layer 1 solution to build elegant Layer 2 solutions on.

Q. Would you elaborate more about a general smart contract execution engine? how can that be achieved any work done on this by Matic?

A. We have some research work ongoing on this, however, it will take more time for us to move forward on it. For example, we are in initial talks with another chain on how to implement Plasma on their platform — we are taking this opportunity to better understand the virtual machine on their platform, and explore different approaches to this.

However, this will take some time, frankly.

Q. Can you please highlight more on the decentralization aspect using your technology. Also in the future, is it possible to quantify the level of decentralization and economic incentivization once mainnet in Q3 goes live or even before??

A. That’s a good question Mihir. We think that Matic Network sidechains will be highly decentralized as compared to many other native blockchains which claim to have high TPS due to Matic’s three-layer consensus model. If I were to create a decentralization spectrum, starting from left where we will have say less decentralized blockchains like NEO, EOS to right extreme being highly decentralized chains like Ethereum/Bitcoin (barring mining pool controversies), I would rate Matic side chains leaning towards the high decentralized score.

Sign up for our Newsletter

More from Polygon blogs