Q&A with Jacob — 18 Dec 2020

Full video: https://www.youtube.com/watch?v=9N4FAV96x1g

In this Q&A session Jacob answers the questions:
1. What is the importance of the Swingby community?
2. What are your future plans after the Chaos MainNet launch?
3. How much volume of BTC/WBTC are you predicting/expecting?
4. Security is a major issue in decentralized finance, how could you ensure that Skybridge will be safe?
5. How does Swingby achieve value capture and long-term value addition?
6. How unique is the technology employed by the Swingby cross-chain swapping?
7. What are the main use cases of Skybridge and what do you think the majority of volume will be sourced from?
8. Besides scalability, the gas fee is another issue for blockchain, does Swingby have a solution for gas?
9. Will you talk to Link for oracles?
10. Who earns more APY, the people pooling liquidity or the people running the Metanode?
11. How many Swingby tokens are required to run a Metanode?
12. Why did you choose to be on two blockchains at the same time?
13. What controls are put in place to protect the $WINGBY tokens being bonded by Metanode during Chaosnet especially when the network is not yet security audited?
14. When can we expect the next bridge?
15. Do you think the markets are looking good?

What is the importance of the Swingby community?

The Swingby community is everything. It’s something that, over the last few months, and certainly while I’ve become a CEO, we’ve been trying really hard to build as much as possible. Inevitably, by this community that we’ve built I mean the guys who are going to take away and start running the Swingby Skybridge, the decentralized network, and the most important thing that we’re building is trying to create this healthy network that people are contributing to so it’s extremely important. And also, it helps because once we launch our Bitcoin bridge, we move on to our next bridge, and we want the community to help out with that as well. Additionally, trying to get attention to the Swingby and show the world why Swingby is such a great product and such a great protocol, the community helps with that too. The meme competition that we run is amazing. We have some really great entries coming into the meme competition. Every time I wake up and connect to my Telegram account, I have very positive emotions, and that’s mostly because of this community. It’s absolutely huge to us, and it’s something that we want to continue building in 2021.

What are your future plans after the Chaos MainNet launch?

Great question. I want to explain a little bit about why Swingby is a bit different from most projects. Most projects have their MainNet launch, they build the hype for this MainNet launch, and once it happened, it’s a step back. The excitement is kind of gone, the product is out there. Swingby doesn’t have that. With Swingby, every single bridge is its own decentralized network, which mainly means that we’re going to have multiple MainNet launches. Right now, we’re doing the MainNet launch to the Bitcoin bridge; in 3-month time, we’re doing a MainNet launch to another bridge and that’s another vehicle of this excitement as well. Whether that’s a Ripple bridge or an ERC-20 bridge, or whether it’s Monero being bridged, that will be another Chaos MainNet launch, and we’ll build excitement to that bridge as well, and we’ll employ the same kind of tactics.

To answer your questions, the main plan after the Chaos MainNet launch is that the first month after the launch, we’re going to focus on the Swingby Skybridge and trying to make sure it’s as healthy as it can possibly be. The ownership of the network is being handed off to the community, we have some significant nodes that are participating, and after that month, we want to start looking to the next opportunity of the bridge. We’re not quite sure which bridge we want to go to next. We have lots of projects reaching out to us, but we’re still not quite sure. I personally would like to see another native token bridge. You got Monero, Ripple, Dash. A Monero bridge is something that I would really love, and other people internally would like to see an ERC-20 bridge. Being able to bridge your DAI or your Uniswap tokens, bridge them to something like Elrond, and make use of the DeFi opportunities. We’re very versatile, and we would really like our community to answer this question, so if you can drop into the live chat right now and mention the bridge that you guys would really like to see, we would appreciate it. Or tweet them to us.

How much volume of BTC/WBTC are you predicting/expecting?

It’s something that is hard to predict. I like to look at some existing products that are pretty similar and the volume they are getting and then try to compare it to what we expect. For example, RenBTC allows BTC to be moved into RenBTC products, the actual use case of them is the most similar to Swingby, and those guys process $5 million worth of BTC to USD value a day. If we expect to receive a modest amount of that, 5% of it, that would be amazing. $250K in the first three months would be an amount we can achieve. We have a lot of strategies in place to make sure we can increase this volume as much as possible, such as maybe some affiliate schemes or maybe work with some more influencers and try to create some explaining videos to show how the bridge works and also integrating it into some existing DeFi platforms.

Security is a major issue in decentralized finance, how could you ensure that Skybridge will be safe?

We’ve written many Swingby improvement proposals explaining how we secure the network and ensure that funds are safe. Firstly, I’ll start with the fundamentals of the Skybridge platform. Basically, the way that Skybridge works is that it’s Proof of Stake, meaning that the more Swingby tokens you bond to your metanode, the more significant your node becomes to the network. The most significant nodes to the networks, like the top 30, have shares of the key. When all of these 30 nodes form a consensus, they can access the funds that are locked in Skybridge and process the swaps. A question might be “what if someone took over the top 30 nodes”. That is pretty much impossible because that would cost significant amounts of Swingby. In order to take over the top 30 nodes in the Skybridge network, you’d have to have millions of Swingby. So, for example, let’s say that there is $100 million worth of Bitcoin locked into the Skybridge, it would always cost three times more to take over the network, so you would have to spend $3 million in Swingby to take over the top 30 nodes. Now, if you have $300 million in Swingby and you affect the network maliciously, the value of those 300 million in Swingby will drop. Now that type of game theory is what protects most Proof of Stake networks and also protects the Swingby network. We have risk mechanisms that essentially monitor the Swingby price, and if the Swingby price drops below the Skybridge capacity, then the decentralized network starts rejecting all kinds of consensus and basically waits for the dust to settle. Those kinds of price feeds come from oracle solutions like Band and Bluzelle. What we’re going to do is combine these two price feeds in order to allow us to implement these risk mechanisms that I’m talking about.

How does Swingby achieve value capture and long-term value addition?

Essentially, the Swingby token exists to power the Swingby network and to operate the Swingby network, and in return, all the fees generated by the Swingby network are given back to its participants. If you want to run a metanode, two thirds of all fees captured from bridging bitcoin are given back to those metanodes. The other third then goes to liquidity pool providers. All fees go back to network participants; none of it comes back to Swingby. Essentially, the value capture is running these metanodes and participating in the network, and in order to do that, you need Swingby tokens.

How unique is the technology employed by the Swingby cross-chain swapping?

We use cutting edge research in MPC and cryptography. It’s a kind of cryptography called threshold signature scheme, and what it basically means is that it’s similar to multisig, it’s like a layer 2 solution of multisig, and you can get a private key split across loads of different people, and when these guys form a consensus and communicate with each other, they can generate a transaction or message. It’s very new technology; not many protocols really use it. A protocol that uses it and is quite popular is Thorchain, as far as I know. We’ve been compared to Thorchain even though our actual use case is very, very different; however, the actual underlying cryptography and technology are kind of similar, so we can be compared a little bit more in that sense to Thorchain. We have some serious cryptography experts that work in Swingby.

What are the main use cases of Skybridge and what do you think the majority of volume will be sourced from?

The main use case of Skybridge right now is going to be this Bitcoin bridge. There’s a $400 billion bitcoin market cap that is the most traded, the most liquid digital asset in the world. On a centralized exchange, the majority of the volume is bitcoin. In the world of DeFi, and we see a huge growth of DeFi, bitcoin can’t really exist, there in a trustless way, it’s almost impossible to get your bitcoin in these decentralized exchanges. It’s certainly impossible to get your bitcoin into WBTC. The main use cases allow people to get bitcoin into WBTC and start using them in this world of DeFi. We honestly think it will help mold the future of DeFi, and DeFi is going to have a massive run, in my opinion. We’ve got loads of cool DeFi projects coming out, lots of really cool other platforms, the elronds, and we’re going to lead the bitcoin charge into that.

Another use case for the Swingby bitcoin Skybridge that I think is going to be pretty big is this concept of liquidity pooling. In the typical DeFi world, you can provide liquidity to these DeFi platforms to get a share of the fees, and Swingby is the same. If you supply your WBTC or bitcoin as liquidity, you earn APR for that, you get a third of all network fees for doing that. The cool thing about Swingby is that you can supply your native bitcoin as well.

Besides scalability, the gas fee is another issue for blockchain, does Swingby have a solution for gas?

A lot of chains coming out are planning to solve this issue of gas fees on the Ethereum network, that holds 99% of DeFi value, 99% of DeFi fund are locked up in Ethereum, and as we all know, that has extremely high gas fees, and using some of these platforms can be very expensive.

We think that if you are a developer and you are building new DeFi products, you are going to be thinking like “I’m just going to build my product on the chain that has all the value, why would you not build it on Ethereum, it has 99% of the value in liquidity”, which is a massive issue because they are competing with Ethereum, which has 99% of the value already. So we think that Skybridge will completely open up the DeFi world and will create a completely open market; developers don’t have to think that if they build their product on Ethereum, it will be locked in, they can freely move between all of these different chains. And that problem of gas fees that we are talking about will be solved by the open market. If another chain will come along that does it slightly better; people can then just move their assets over there, execute their trades and come back to Ethereum if they want.

Swingby is not going to directly solve the gas problem, but the value of Swingby existing in the ecosystem and how we want to change the world of DeFi will absolutely solve that problem.

Will you talk to Link for oracles?

Regarding oracles, we are happy to partner with as many as possible; we’ve got BandProtocol and Bluezelle, which are both huge players, but the more oracle solutions you have, the more you can aggregate these feeds together, the more secure it becomes, and the less redundancy.

We are going to be partnering with more oracle solutions.

Who earns more APY, the people pooling liquidity or the people running the Metanode?

The answer is the people running the Metanode, they get 2/3 of the network fees, and the liquidity poolers get 1/3 shared between them. The reason for that is that Metanodes are in charge of security, and for securing the network, they should have the bigger share; you have to pay money to run these nodes, that is why they receive a bit more of the fees.

How many Swingby tokens are required to run a Metanode?

150000 Swingby tokens will be the minimum for running a Metanode. The more Swingby tokens you bond to your Metanode, the more of the network share you receive.

So 150000 is the minimum, and you can earn exponentially more if you put more Swingby tokens inside that Metanode.

The reason why that is quite high is another security mechanism that we have to consider with Swingby. As we have these liquidity pools that the Skybridge nodes control, the amount of value in these liquidity pools can never be higher than running the metanodes in the network, so basically, we need to try and incentivize as many Swingby tokens as possible to be pulled into these top nodes in order to improve the security because the more value that is in these metanodes, the more capacity we can store in the liquidity pools.

Can you explain THORChain and Swingby?

When Swingby launched, we were compared quite heavily to THORChain, and the reason for this is that the underlying cryptography that we use is quite similar, the threshold signature scheme, the Binance SDK. That is why we are compared with THORChain, but the actual use case of Swingby is very different.

What THORChain is trying to do is to run their own kind of chain ecosystem with their own AMNs and decentralized exchanges and then bridge other chains into THORChain.

Swingby has a completely different goal; we don’t even want to have our own AMNs. What we want to do is just connect other chains together, we are not interested in siloing our own ecosystem and absorbing money into it, we just want to bring value to the network and not compete with liquidity. The use case is very different.

Will the Swingby foundation dump tokens?

It is very normal in the world of Crypto for the foundation to sell tokens in order to cover costs and that kind of thing. Swingby, which is different than most protocols and most projects out there, is very transparent with you on when we sell tokens, how much we sell, what price we sell them for and our current balance, when they unlock, etc.

We release these monthly reports on Twitter in which we say how much we sold, what price we sold, and so on.

Any plans to list on another exchange to increase liquidity?

Absolutely, one of our main goals is to provide more liquidity venues for the Swingby token. I can’t talk about who they are or who we’re in contact with because most of these exchanges do not like that.

One of the other strategies of Swingby right now is to focus heavily on one specific market and try to make sure that it has all the liquidity because more often than not, when you open a new market somewhere, sometimes it doesn’t even provide more liquidity, it just spreads the liquidity to other markets and that causes further issues because it becomes more difficult to buy or sell your Swingby token on other exchanges.

So our strategy is to focus on one market, and when another big exchange comes along, we will go full throttle on that one and try to provide as much liquidity there.

The Swingby token is quite liquid at the moment. Over the last few months, we have seen massive rises in liquidity through the excitement of Chaos Mainnet launch; right now, it is quite easy to start liquidating $100k worth of Swingby, something that a couple of months ago would have been a completely different story.

I think that going into the Chaos Mainnet launch, when the product becomes more usable and when we have these Metanodes and reasons for people to buy the Swingby token and earn passive income from it, we are going to see a lot more liquidity coming in, that is my theory.

Will there be a small amount of burning when tokens are swapped?

We employ lots of different mechanisms on how the Swingby network will take some of its profits and incentivize the token price to go up in value. We are already doing some token burns; we announced a 10% of the supply burn, which is going to be happening from December 28th, which is the launch date of the Chaos Mainnet, onwards.

On the 28th of December, we will burn 25% of that 10%, so 25 million Swingby tokens will be gone forever, and then every single week, we will take the volume of the Bitcoin bridge, and we will burn that in Swingby, we will keep doing that until 10% of the supply is gone forever or 8 weeks have passed.

A lot of people think that burning is just a mechanism to pump the price of a token, but it is not actually; it drastically improves the health of the network in a few different ways. The first way is that it adds a constant deflationary value to the Swingby token, meaning the amount of tokens existing is slowly decreasing, and that reduces the drawdown potential of the token price; it means that if tokens are constantly being burned, big volatile movements down in token price are less likely, which is beneficial to people running metanodes. If you have bonded Swingby tokens to your Metanode, you are less likely to start thinking “I’m going to pull my tokens out because there is less drawdown potential that the price is increasing”.

Token burning also adds an appreciation to the token price, so the actual token value should, in theory, increase significantly with the burning; therefore, the capacity of the network will also increase because if the amount of Swingby tokens bonded into these Metanodes increases, if the value of the tokens increases, the amount of liquidity pulled in the Skybridge can also increase, and therefore, there’s more swap capacity.

Token burning will also incentivize people to actually use the Bitcoin Skybridge Network and swap more, and that increases the overall health of the network, more fees are being paid to Metanode operators and liquidity pool providers, and it is also nice marketing.

We don’t want to burn too many Swingby tokens for a few reasons. Firstly, these tokens are required for maintaining the network, and we are only on the first Bitcoin bridge now, but we are expecting to have two or three Bitcoin bridges by the end of the year and potentially more, and every time we launch a new bridge, more tokens are required to be locked up, so we don’t want to get rid of too many. Secondly, we want to do it in a nice and healthy way, the economics of the Skybridge network are very sensitive.

Why is the majority of your team anonymous?

In cryptocurrency, it is very usual to have anonymous teams; we use this thing called “the $5 crowbar analogy”, and the concept is that you can have the best security in the world to protect the network, but if someone can just buy a $5 crowbar from the gas station and come over and just beat you with it until you give them money, then your security is useless, so that is why a lot of crypto projects like to remain anonymous and decentralized because, for example, if I get beaten up with a crowbar, we still have the development going on, we still have the foundation operating in a normal way.

I came out as CEO because I wanted to try and take the best approach. I understand why people like to see a face behind the projects. I stood in and said that we need to be more public about this; the rest of the team can stay anonymous, but I came out trying to build confidence in people.

So that is why the majority of the team is anonymous. We do have a lot of other public figures, but the development team is specifically anonymous.

For Swingby to be as realistic as you say it needs a strong, experienced team, so can you recommend your team (Founders, R&D, Developers, etc…)?

Many of our team members are anonymous, but we do have an extremely experienced team. I have worked with Bitfinex, and I’ve worked in cryptocurrency for the last four years. Our CTO is from another major exchange and has over six years of experience. We have many people from top companies working on this, and you can see that in the quality of the technology that we are bringing. The technology of Skybridge is so complex, we are venturing into completely new territories, and we have got an experienced team backing them.

What is the real difference between a DApp running on Ethereum and DApp running on SWINGBY Network? What’s the benefit of one approach vs. the other?

A: We don’t allow DApps to run on our network. We are not a lot of utility in that sense; we are not like Ethereum or anything else. Essentially, what we do is just building bridges from one network to another, we allow smart contracts to be used on multiple ecosystems, so you could build a smart contract on Ethereum but also have it communicate on the Elrond chain and stuff like that.

Why is Swingby building on BEP2 (and going to be BEP20) and not ERC-20?

A: When we started building the Swingby Protocol, the bridge that we wanted to build was something like Bitcoin to Binance chain, but we think that the environment has changed quite significantly since then, and we realized that the main thing that people want is Bitcoin on Ethereum, that is why we are approaching that. The Binance chain is still on our spectrum, and it could be the next thing that we do after the Bitcoin-to-Ethereum bridge.

That is a classic case of crypto moving so quickly and having to evolve, picking the most popular bridge that solves the problem that needs solving the most, and that is Bitcoin.

TOKENOMICS is very important. What are you going to do for investors to HODL longer?

A: Tokenomics is extremely important, but I think that what is important to note is that we build a very healthy platform, a very healthy network that has extreme amounts of usability and has loads of people using it and high APR paid out to the network participants, and the token price is going to improve because of that. I personally dislike it when protocols and projects just focus on raising the token price. If they are doing that, it is a bad investment; those guys are probably just trying to grab as much money as possible. That is not how cryptocurrency is founded; it’s not how it should work.

Cryptocurrency companies should be focusing on their network and try to build a healthy network that indirectly brings value to the token; that is what I think that Swingby will do.

What are the minimum hardware requirements for running a Metanode?

This is something that we have been testing in the last few days, trying to play around with various kinds of hardware sizes and trying to see what the minimum requirements are and how well the Swingby network operates on. We have come to the conclusion that the absolute minimum for a Metanode is 4 CPUs or VCPUs, around 8 GB of memory, and about 600 to 800 GB of storage, and those are like the absolute minimum.

The reason we require a bit more powerful machines is because of the concept of threshold signature, where I mention these multiple nodes that come together and form a consensus and then sign the outbound transactions. The more parties that are communicating and forming this consensus, the more resource-intensive the threshold signature becomes. When we release the first Skybridge, we are going to be setting it to around a threshold of 16 or maybe 31, and if we set it higher, we will require more powerful machines.

We require a lot of storage because if you think about the Skybridge network, it actually uses two chains, so you need the Bitcoin node, the Ethereum node, a Bitcoin BlockBook indexer, so you also need an Ethereum BlockBook indexer, and because of all that, you need quite a lot of storage.

Pricing wise, we have been kind of shopping around to identify the best places to host this kind of instances and quite like Scaleway. If you provision an instance with these kinds of specs, it will probably cost you around $40–50/month, which is pretty cheap.

Since the requirement for a Metanode is quite high, will there be any delegation of Swingby tokens to active metanodes?

There won’t be, Swingby does not support delegation, and that is because we are a layer two protocol essentially, we don’t have our own blockchain, we rely on other blockchains, and we just connect them. For this reason, we can’t employ things like slashing.

In the future, we will be increasing the number of nodes and maybe even adjust the minimum cost to run a Metanode.

How do you calculate how much APR the Metanode operators will receive?

It is very difficult. Basically, there’s a lot of variables in the network. Essentially, what we did was try and model this as best as we can; we created a model expecting like 50 nodes in the network, and each of those nodes has the same amount of Swingby bond. The second you start bonding more to one node, you start earning more than the others, and it kind of skews based on that. So we wanted to try and work it out as if every single node had the same amount. Obviously, that is not going to be the case. You might have a node that has a lot of tokens, and you might have a few nodes that don’t have as many tokens, so obviously the APR is going to be skewed based on how many tokens you bond to your node.

The model is as accurate as possible. We took into pre-consideration that every node would have the same amount, and based on the amount of volume, the APR would increase or decrease significantly. For example, if we hit our target of $250,000 worth of volume daily, these metanodes will be earning around 40–50% APR a year if they have the same bond. If we, for example, go to the $1 million mark, then the APR paid out is exponentially increasing again. So the actual APR paid to your node is infinite, in a sense, realistically, we would say around 60–80%, but it is quite hard to model.

No matter how good a product is, it dies without marketing, any plans for massive marketing?

In the sense of crypto, I don’t think that is entirely true. In cryptocurrency, I think that the projects that focus on building healthy networks and products are the ones that become the top page coin market cap products, the billion-dollar market cap platforms. I think that is because a lot of companies that just invest all the money into marketing and stuff like that are planning on an exit, they are planning for their protocol to die, whereas the projects that invest mostly in building a healthy network are the ones that inevitably succeed, and you can see that in things like Ethereum, Bitcoin, and all those protocols that have solved actual use case, they are going to exist and people are going to use them.

Obviously, it is important, and it is something we are spending money on. We have been spending a lot of money on it over the last few months, you can kind of see it with the token price, which is up around 100%, and we will continue to do that. I think that Swingby is a very marketing-friendly protocol because we have constant Mainnet launches, we have the Skybridge Mainnet with the Bitcoin bridge launch now, then in a few months, we are going to be launching the Mainnet of the next bridge, and so on, so it is a very-nice-to-market kind of protocol.

We are going to be working with a lot of influencers to make more videos and talk about it more. We are going to be doing more things like this Q&A to give the community a chance to ask questions and to answer these questions.

Any plans to integrate with wallets that connect Dexter’s directly, like how Celsiuns and Trust Wallet connect to?

Because Swingby is layer two, which essentially means that you don’t use smart contracts, we kind of operate above the blockchains. You can actually use the Skybridge network with any existing wallet because you don’t need to interact with a smart contract. Swingby shows you a deposit address, you send your Bitcoins into it, the Skybridge then processes it into WBTC and sends it to an Ethereum address that you told it to send it to, which basically means that every single wallet in existence that allows you to hold Bitcoin or Ethereum works with Swingby’s Skybridge, and that is a massive advantage over a lot of projects.

We also want to integrate directly with some wallets, we have got some communication channels open for that, basic straight integration, and we are also building a real widget you can embed into any website and any app, and the second you embed it, it will allow you to start bridging Bitcoin into Ethereum and vice versa. That is one of my main strategies as the CEO, going forward and speaking with a lot of other projects trying to get that integration stream going.

Why do you need to run Metanode to confirm transactions? Running Metanodes need the cloud makes me envision it working as PoW protocol? Why don’t we validate transactions through node staking (PoS)?

The way Swingby works, if you actually break it down, is essentially like a private key that splits out into loads of different shares and whenever coins go into that private key or that address associated with the private key, the network processes an outbound transaction, so when Bitcoin comes into it on one side, WBTC goes out on the other side and vice versa. You have to employ this kind of proof of stake mechanism and characteristics in order to trust people in having a share of these keys. If you have a lot of Swingby tokens, you are not going to cause damage to the network; therefore, you can trust them more, and that is why they get a share of the key. That is kind of how the concept of Metanodes exists, all the Metanodes basically control a share of a private key, and when they witness a transaction going into their address, all the metanodes at the same time agree to form a consensus and sign the outbound transaction that we talked about.

Why did you choose to be on two blockchains at the same time?

Back when we started, our main goal was to solve the problem of bringing Bitcoin to the Binance Chain, but we kind of changed the environment since then. It is unfortunate, but it is just something that happened. It doesn’t cause any additional damage to the network, it just means that Binance Chain tokens are just not ERC20 tokens. One of the annoying things is that I would have loved to have Swingby in something like a Uniswap pool, but we can’t do that. If we build a bridge from Binance Chain to Ethereum, we can just bridge Swingby into ERC20 tokens, that is something that I personally would like to do too.

What controls are put in place to protect the $WINGBY tokens being bonded by Metanode during Chaosnet especially when the network is not yet security audited?

Your Swingby tokens bonded into a Metanode are under absolutely zero risk. There is no risk of slashing. You even maintain custody of the private key of these bonded tokens, and nobody has any kind of interaction with these tokens. There is no way that you can lose your tokens at all. If you have some Swingby tokens, you send them to yourself, and you specify your peer-to-peer Metanode key in the memo of that transaction, and then you time lock them for 2–3 months. Once they are time-locked and you have specified your Metanode peer-to-peer key in the memo field, then that is considered by the network as a bond, so you don’t lose custody of those tokens, so there is zero risk.

When can we expect the next bridge?

The next bridge will be at the end of Q1 2021; at the end of March is when we have confirmed that the next bridge will be live. The second that Bitcoin bridge is live, we are straight away focusing on the next bridge and building excitement for that one.

Do you think the markets are looking good?

Yes! In the broader landscape, you’ve got Bitcoin setting a new ATH, which is absolutely huge, you’ve got all this institutional money coming in, cryptocurrency, in general, is looking amazing. We’ve got 2017 vibes; it is a great time to be in the ecosystem. The DeFi perpetual index, which is basically a Futures contract that tracks the top leading DeFi projects, is also looking good, it is on a solid uptrend, and that basically means that there is a lot of excitement coming in around DeFi again, and I think that is going to continue definitely into 2021.

I think that as we approach the Ethereum 2.0 launch, that is going to significantly pump. There is going to be a huge amount of interest coming to DeFi, and that is why we are positioning ourselves as this DeFi protocol.

There couldn’t be a better time to be Swingby and to be bridging Bitcoin to WBTC.

We haven’t even launched the Chaos Mainnet yet, so I think that the second Chaos Mainnet does launch and people understand the problem that it solves and start using it, we are going to see far more money coming into the market, so I’m absolutely pleased with how the markets are going right now.

Want to trade Swingby? Check us out on CoinGecko or CoinMarketCap.

Get Rewards! Stake your Swingby Token to earn up to 52% annualized (67% compounded) in rewards, airdropped to your wallet weekly. Use the Staking Portal or BitMax.

TestNet Explorer: bridge-testnet.swingby.network/explorer
Staking Program: https://swingby.network/earn

Telegram: https://swingby.network/telegram
Twitter: https://swingby.network/twitter
Reddit: https://swingby.network/reddit
Discord: https://swingby.network/discord
Github: https://swingby.network/github

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store