Bancor DAO Discussion #3 (November 30, 2022): Carbon Q&A
The following excerpts are paraphrased from the recent Bancor DAO discussion on Discord.
(2:13) Intro discussing Jen’s post on Discourse
Bancor is becoming the umbrella for a suite of new products. Carbon, the first new product, can play a major role in generating fees to restore reserves and reduce the deficit in Bancor v2.1 and v3. At the same time, Carbon can chart a promising path forward for the Bancor DAO with its “asymmetric liquidity” model.
Read Jen’s post in Discourse: Carbon: An update to the Bancor DAO and Community
Also a summary was posted of the Carbon DEX Twitter Spaces #1.
(10:03) Range Orders in Carbon
Your order, when you set it, is visible for everyone who is on the blockchain to interact with whenever they feel like interacting with it. When you’re setting a price range it has the same kind of hyperbolic behavior that we’re already familiar with. Selling between $2,100 and $2,000, it will sell at every single price between those ranges until it has sold everything, precisely the type of behavior we would expect from a concentrated liquidity curve. It’s not the case that it sells 5% in small increments 20 times. It’s continuous pricing the same way AMMs continuously price anything. It’s that same idea, it’s just to an infinite number of small increments so you get complete price continuity. It’s not the case that the protocol goes and sells the tokens you have nominated. Rather, it is advertising the price that you have chosen. In a way, it becomes sort of a request for quote protocol.
(20:02) Carbon Range Orders vs. traditional Range Orders:
When you set up a limit order in a traditional context, you just set up a single price. If the market happens to reach the price you have requested, someone is going to take you at that price, and you’ll get the transaction that you requested. Traders set up dozens of these limit orders in ranges because they have no idea which way or how far the market will move. Carbon is an order book that was designed to allow that type of behavior. So rather than have to set many many limit orders one after the other at different price ranges, instead you can just set the price range that you want to execute your strategy over.
(22:12) Re-usable liquidity in Carbon trading strategies
The other important thing, after a limit order is fulfilled, or even partially fulfilled, the fruits of that sale would just be returned to the user’s account. If you want to use those funds to buy something else when the price recapitulates, and this is common for people channel trading or range trading, then you need to wait for that limit order to be filled, wait for the funds to appear in your account, and then use those funds to set up another limit order in the reverse direction…On Carbon, you can actually come into the protocol with that type of mentality, that premeditated strategy where you have conviction about which way the markets are headed, and that you have a very well defined risk profile that you’re happy to tolerate, and you’ve defined your price ranges and after tokens are being sold for another token, then that token immediately funds a reverse trade at a different price point. This is exactly how professional traders make a living. To give traders a tool that is specifically purpose-built to facilitate that type of behavior I think is something to be really excited about. It’s an order book, yes, but it’s not just an order book. It’s an order book like you’ve never seen before. It’s an order book that is built specifically to enable traders to do the stuff that they’re already doing but in a way that is a lot more frictionless and a lot more intuitive.
(25:25) Is Carbon an AMM or an order book system, or both?
I do see them as one in the same. It’s not so clear cut. Certainly you wouldn’t call a traditional AMM an order book. Then you’ve got this next generation of protocols starting to look a little more order book like. AMMs started life as being an alternative to order books so we could have continuous on-chain liquidity; that it has unlimited liquidity at all price points. It certainly has a place in DeFi, and we’re going to have to learn how to merge these ideas together. But we know what the problems are with traditional constant product AMMs. There is reason why these more order book-like products are gaining so much traction. It’s because the active traders, the people that actually move volume around on blockchains, that’s the kind of service, the kind of product that they want to interact with. Would I call it an AMM? Yes. An AMM is still in my mind just a series of smart contracts that facilitates an exchange of assets in a decentralized manner between anonymous users. So in that sense, of course. It’s still an AMM and it still uses an invariant function to automatically price assets along a bonding curve. But it’s, I think, the first example we have of a non prescriptive AMM. An AMM that says you choose the price and you choose the algorithm, rather than the AMM chooses it for you and you choose whether or not to participate. It still has the same bones of an AMM, but it looks and behaves very differently. It is still an AMM, but it is also an order book, and you don’t need to make a distinction between these things.
(29:46) Why DeFi needs more traders
The most healthy markets are the ones where stuff is being traded all the time. Not just being hodled, and dumped all at once, or liquidated in a rush and causing all these cascading events. We do have liquidity problems across all of cryptocurrency, and I think the reason we do is bc we’re still reliant on these kinds of AMM safety nets that kind of bring some sort of liquidity support to these assets. But we also know because of how predictable they are and because of how they are abused by the amalgam of everyone in the industry. When there’s panic, and a need to get rid of a lot of tokens, AMMs become the punching bag of the industry’s frustration. It’s liquidity providers in those protocols that end up having to weather a lot of that injury. I think we don’t necessarily need that to be true. I think it would be much better to have a system or to have a culture or infrastructure that supports people that are willingly buying assets at cheap prices because that’s what their strategy dictated they should do. If the price of something is quickly accelerating it’s because people are willingly selling tokens at those prices. And they are happy to have taken profits at that range rather than liquidity providers who are left wondering what happened to all the tokens they had that they didn’t necessarily have ambition to sell at that range. This is important. An AMM is still kind of an automatic orderbook. A lot of people in DeFi and using liquidity products neglected to realize that they are actively trading the tokens that they provide. Having an active protocol that actually gets users to declare what prices they want to buy and sell at is just much better for the user experience overall because they don’t have that sense of being manipulated or that sense of having decisions made on their behalf. This is why the industry needs more day traders. We want people who are agreeing to buy and sell stuff at all kinds of prices all the time keeping the industry liquid rather than a passive liquidity base who doesn’t understand what it is they are doing exactly.
(36:51) Every single Carbon order is its own AMM
In Carbon, you don’t pool your assets with everyone. Every single trader is using their own bonding curve. You are in a pool with no one but yourself. There is no one participating with you. You are an island of your own liquidity with very clear distinctions between each user. Each user has their own bonding curve. You are only impacted by your own actions. The actions of everyone else does not have an effect on you.
Q: Is there a risk whales could cause a panic or a bank run in Carbon?
It’s worth considering what would a bank run even look like in this case. There’s no advantage to being the first one out. Whales can move the market, and of course they will. Day traders are not hoping that the whales don’t do something. They’re actually betting on what the whales are going to do. You don’t have to agree to sell something anymore if you don’t want to because you set predefine specific price ranges. If there is a big whale movement that moves the price of something down, if you’ve set an order to buy at that low price, then all that’s happening is your order is fulfilled at your request. If you’ve got an order to sell ETH at $20,000, and a whale moves the price to $20,000, then so be it. You’re a passenger in what these forces are doing with the valuations of stuff. The question is now that you’re aware of these things, how are you going to trade with it? And the answer with regards to this product is to give you the capacity to act on those instincts rather than have to sit by passively and wait for the market to move you around.
(53:40) Carbon innovations could be used in a token launchpad
If there is a real appetite for launching legitimate tokens in a decentralized way and providing those early adopters of a token some sort of assurance that the funds being raised are going to be used for legitimate project building purposes then I might’ve expected certain token launchers to be more popular than they were. Whether or not this is something that the DAO expects to see or wants to see will be an ongoing discussion. The first time that I had the Carbon equations in my hand, the first thing I did was explain a token launcher. The reason is because if I was to participate in a token launch, there are certain things I would want to see. One is some kind of schedule as to how price discovery for a token might look. They tend to follow a sort of dutch auction scheme. What I would prefer is a token that launches at a stable price, that’s predictable. So we can do that. The Carbon equations allow for what’s called constant pricing. We have a parameter in the equation that allows the curve to be infinitely amplified. This means zero slippage and zero price discovery. But you can’t have this forever. Eventually you need regular market forces to take sway over how a token is priced. On Carbon one of the things we can do is slowly dial back the token amplification from infinity all the way back to constant product. You can actually tell people ahead of time how slippage will likely impact the ability to trade this token on this schedule. Early adopters, if you buy a whole bunch of tokens, and you want to get rid of them because you get cold feet, you can give them all back without the risk of anyone front running you or rugging you or anything like that. If you have conviction though, the tokens price discovery can slowly be dialed in and over time the market can make up its mind about what it’s worth as it becomes established. And I think that that actually sounds like a much better idea than trying to rush it all and launch everything on a constant product curve from day 1. I’m convinced that in time everything will go back to normal, and there will be new projects that will be looking for ways to delineate themselves from rugpull scams, and I think Carbon is a good way to do it.
Reach out to Jen if you’d like to see the Carbon token launcher powerpoint.
(1:01:15) How exactly can fees from Carbon be used to restore the Bancor v2.1/v3 pools in deficit?
Jen: “I’m an LP on v3, a LINK LP, and personally support Carbon fees going directly towards alleviating the deficit. I can’t say for sure, because it’s ultimately a DAO decision, but as a member of the DAO, I want to speak out and show support. The entire reason this new product was created was to help alleviate the deficit, and I’d like to see that followed through.”
Mark: “I also have my LINK on v3. When I say things like Carbon was designed specifically to create an augmented revenue stream, something that is successful enough, something that has the potential to be popular enough and used enough, that it can simply pay down that deficit, I mean that not in the sense that I’m white knighting for the protocol, I’m also selfishly wanting Carbon to be successful enough to do that because my own non BNT tokens are in the protocol along with everyone else’s. We’re not necessarily limited to certain mechanics…This is something we need to explore more thoroughly with the DAO. I think it is important to have a functioning product, something we can all rally behind and believe in, something that is successful and makes sense and is addressing a market need first before we start looking at these mechanics. It’s still in some way, predicated on the success of the protocol overall. I’m convinced this type of limit order system, and the types of sophisticated players and volume and flexibility that it brings to the industry is going to be that thing that we’re looking for.”
…If you’ve chosen to remain on v2.1, don’t worry. The way that the deficit will be addressed will need to necessarily be spread over v2.1 and v3. So if you’ve chosen not to migrate, don’t feel like you’ve missed the bus. We wouldn’t do that. Whatever success we garner for Carbon, however long that takes, and whatever needs to be done to realize that goal, rest assured that if you’re in v2.1 you’ll still reap the benefit. You don’t have to migrate to v3 for that purpose. It would be ethically and morally apprehensive if Carbon only benefited the people that migrated to v3. Everyone who is in the legacy protocol is in it together, and Carbon is meant to be here to benefit all of us.”
(1:17:40) Carbon progress so far
Before the whitepaper, before the litepaper, before Carbon had a name and so on, the smart contract developers were already working on how to realize these features and make it compatible with the EVM. Really it’s been in some state of active development since the end of July. When you give users this kind of flexibility, when you afford them the agency to choose their prices, their token amounts, and have it be not a part of some gigantic liquidity pool, but for each user to be their own individual trader, to have their own specific order book profile, their own specific position and executing their own strategy, that kind of complexity has to be abstracted away into something. This means things like order matching and routing through the AMM is a lot more complicated in something like Carbon than it is on a traditional AMM. Those kinds of problems we really needed to sink our teeth into and make sure it wasn’t an insurmountable obstacle before we started presenting this stuff. So by the time the rest of the world found out what it was that was being worked on, we had already gone through the motions of creating the first draft of the smart contracts and seeing that we could achieve the type of behavior that we wanted. There are smart contracts already that do what we want them to do. It’s in the phase now of essentially ironing everything out and making sure that it’s as efficient as it can be. Obviously we need to get into code audits, and work out the UI, what that’s going to look like and how that’s going to behave, but also connecting with other people in the industry and making sure that they understand what these smart contracts do and how to use them. Often they want to create their own smart contract layer on top of it. It’s not even that we’re at the stage where we’ve got our own smart contracts ready, but we’ve already started speaking with other residents in the DeFi space that may want to build their own interaction layer on top of it. It’s well and truly underway. I would say we’re way past the halfway mark, and we’re in a more detail oriented phase where it’s starting to wonder what the app is going to look like and what needs to be done to make it the most intuitive and enjoyable experience for everyone that wants to interact with it. I’m not committing to deadlines, but as far as the development process, the longest and hardest part of it is already behind us.
(1:25:28) Trading strategies can be made with any standard ERC20 token (including non-BNT pairs)
Orders can be created with any token that you want. It is no longer required that every token be paired with BNT. The protocol fees are collected exclusively in the tokens that are being made and traded. In that sense, the protocol should collect a pretty diversified repertoire of different assets, and it’ll be up to the DAO to decide fee parameters. Obviously alleviating the deficits on v2.1 and v3 are paramount. There’s absolutely no question in my mind, and I think you’d be hard pressed to find anyone in our DAO that doesn’t see that as the most urgent priority use of those funds. After that, this could persist as a type of treasury system that is composed of literally every token that is traded across the protocol, and in that sense, I think it should become a little bit more antifragile. Something everyone should consider and take to heart.
(1:27:41) Gas efficiency of Carbon
The benchmark that I’m setting is that it should be equal to or less than interacting with Uniswap v3. Certainly in terms of strategy creation I think it’s going to be a lot less. As someone who’s creating orders it should be very very cheap. As a taker, it can be ridiculously cheap if you’re only interacting with one position. We have a working prototype, but this is by no means complete. This is something I want to keep researching and keep iterating on to make it better and better. The gas overheads in general are looking significantly better than they ever have, precisely because the simplicity of the order system is actually pretty good. There’s nothing really to exploit in terms of the normal shenanigans that people play against AMMs. Things like throwing the TWAP price by conducting very very large trades in one direction or the other direction in order to force liquidations and that kind of thing, or affect somebody else’s liquidity provision, or front run someone else’s trade, all these protections and kinds of considerations that you need to build into an AMM just don’t apply here. Even things like sandwich attacks won’t work on Carbon because of the way that the asymmetric liquidity works. By design we’ve put in the extra work to make sure that the equations by themselves perform a lot of that extra work so that we don’t need to add a huge amount of additional logic and things on top of the smart contracts in order to facilitate that type of behavior. I’m still very very optimistic, and I think you’ll be pleasantly surprised at how lightweight these interactions will be.
