The CEO of Radix DLT, Piers Ridyard, recently conducted a Q&A session with the 17,000 member @CryptoDiffer Telegram community. These Q&A sessions have proved to be an excellent way for Radix to connect with the broader crypto community, and some excellent questions were put forward and answered by Piers.
The full transcript of the Q&A session is below.
Q. Can you introduce yourself to our community?
A. I got started in crypto back in 2015, when I started mining on the genesis block of Ethereum. After that, I started a blockchain company that went through YCombinator, after which I exited to become CEO of Radix. Dan, the founder of Radix, has been in crypto since 2012. A real OG :-D
Q. Can you introduce Radix to us?
A. Radix is a layer 1 protocol built specifically to serve DeFi. We like to call it layer 1 DeFi done right. That is because Radix is the only decentralized network where developers will be able to build quickly without the constant threat of exploits and hacks, where every improvement will get rewarded, and where scale will never be a bottleneck.
Q. Let`s now talk about the milestones you have achieved so far and about your upcoming plans?
A. We concentrate on delivering three core innovations to the market - a way to 10x DeFi developer productivity, a way to 100x DeFi execution efficiency and security, and a way to make sure gas fees/transaction fees always stay low, no matter the scale the network gets to.
Q. Sybil protection. I recently revisited the consensus roadmap on github. I seem to remember it used to say PoS and/or DPoS for Sybil protection for RPN-2+. Now it only says DPoS?
A. Good question - Radix is both POS and dPOS, which is why it is confusing. Nodes can stake to themselves (and that is visible separately) as well as other people being able to also stake to those Nodes. On Olympia (our first network) there is a maximum of 100 node runners that can be participating in consensus in any give epoch (however, more can run nodes if they want to). In Babylon, we expect to open up this number and then at Xi'an for the number of participating nodes in the network to essentially be unbounded.Radix is both dPOS and POS because you do not have to run a node to be able to get a share of the staking rewards, but you can also run a node if you choose to do so. We think it is the best of both worlds.
Q.Is it true that we only need 100 validators to service all these private universes and the public network, or does each private universe need its own set of validators?
A. In a private universe, you can set up as many or few validators as you like. 100 is the number we chose for the first Radix mainnet, but it could be more or less. For future versions of the Radix mainnet, it will be much more.
Q. Is there an impact on dApps created on RPN-1 when the transition to RPN-2 occurs? No refactor needed?
A. No re-factor of what is built on the ledger, but it is highly likely that the Node APIs will change, so any integrations will need refactoring. This is because the Radix Engine is being upgraded from v1 to v2. All on ledger state will be preserved, but calls to the ledger will likely change if you are building things like wallets etc.
Q. Is there a way to calculate the demand to TPS ratio? How do you measure current DeFi demand?
A. haha - very, very good question. No, there are currently no benchmarks to actually work out how to describe DeFi demand. Transactions per second are not really valid because the difference between swapping on UniSwap and adding liquidity to Uniswap is vast from a complexity/ledger throughput point of view (hence the huge difference in gas fees)
We are thinking about better ways to do benchmarking for stuff like this, but it is still early days. I don't think the market fully understands yet that TPS numbers are not the important thing; DeFi operations are the critical measure, but so far there is no real standard for measuring that effectively.
Q. Radix Application Api has been deprecated (reference GitHub radixdlt ) . Does that mean new version of api's would be available in Olympia ? Or would they remain functional and will be removed till Rev2 comes along the way?
A. Everything is going to be just JSON RPC for the time being. We are keeping it light and simple; there will be some new documentation coming with Stokenet. This isn't yet the transition from Radix Engine v1 to v2, which will start at Alexandria and be completed with Babylon.
Q. I think it is true that you will always need some maximum number of validators per shard group. Eventually, this maximum may be more than 100, but some limit will always be required (if only to reduce the storage/CPU requirements for the once per epoch software. Do you agree? Or does your hybrid combinatorial POW being tested in Cassandra make possible an unlimited number of validators per shard group?
A. This is still being researched. Cassandra is essentially leaderless and allows emergent ordering for consensus operations. This would mean that a maximum 100 validator quorum per shard could be dropped. Some good potential here but certainly not at a stage that I could say with certainty.
Q. Are the advantages of Blueprints so high in cost-saving (dev work (basically just the copy-paste in this case) + security auditing) that the team thinks this behavior (the workaround) will not occur?
A. Good questions. For context, Radix Blueprints allow a developer to create the template for a type of smart contract that anyone can call and instantiate as a component on the Radix ledger, massively saving on time and cost for building and deploying dApps. For example, someone could create a CFMM Blueprint and get it security audited. Then any entrepreneur could come along and call the Blueprint to create their own CFMM on Radix (e.g. Uniswap, Suishi swap, Pancake swap etc).
Every time this happens, the developer gets a royalty both for the instantiation and for the use. The developer gets to choose what royalties are charged for the instantiation and the use (including making it free).
It's a little like the difference between using a supported code library vs an unsupported code library - most people prefer to just use a supported code library because it is simpler. Cost is only one component; the time (for example, to get your copy re-audited) and distraction from just getting on and delivering are all things to think about.
Sure, some developers will choose just to copy other people’s work and not use Blueprints, but that will increase their own technical debt and support burden, plus they do not get the benefit of a security audit against their Blueprint, which they would have to re-do. I expect both will happen, we think most people are going to highly value speed and ease of just instantiating from existing useful Blueprints rather than working around them. At the end of the day, it is a free market, and people can choose whatever they want to do.
Q. Will it become multi-fold more complex to delegate when state sharding is active which brings in more validator groups?
A. No, there will always be a single staking shard where all the nodes have to register and this keeps track of all staking and unstaking as well as slashing. This was the one concession to global state that we made on the sharded Radix network. Staking and unstaking has sufficiently low volume throughput even at millions of nodes that we didn't see it as an immediate issue.
Q. How does Cerberus deal with the risk of double usage of a shard with massive (independent) parallelization of processing of transactions?
A. All transactions/operations are deterministically mapped to shards. Double usage of a shard is not an issue, a shard can contain many more than one transaction. If two transactions are on the same shard, and the state transition is requested in parallel for both of them, and they conflict, then you will get a state conflict on the quorum leader for that shard, for that consensus round, which will choose which of the two it does first, then discard the second.
However, the probability of that happening will be 2^256*2^256 (not considering the parallel request if not forced). It is very, very difficult to achieve two separate operations on the same shard by force. So, I would be surprised if I saw it happen in my lifetime at 1m TPS per second from now until I die. If it does, the network can deal with it via the Cerberus.
Q. Can you please confirm that stake weight along with other factors including availability, responsiveness, optimal latency are considered for choosing a node in validator set ?
A. Short answer - no, just stake weight.
Long answer - yes because poor-performing nodes get punished by the protocol for not participating in consensus rounds (they don't get their share of the emission). This means that while the network doesn't directly choose well performing nodes, there is a market mechanism to make sure that well performing, responsive, reliable nodes with high uptime are MUCH more likely to be picked by rational stakers.
Q. How will the transition between RPN-1, RPN-2, RPN-3+ be managed? Is there documentation available online explaining this?
A. Documentation will be provided closer to the time on how the transition from Olympia to Bablyon to Xi'an will happen.
Q. Are smart contracts coming with Alexandria or is it until Olympia?
A. Neither - smart contracts come with Babylon (2022). Alexandria (Q4 2021) is when we launch our new smart contract language: Scrypto. We will then have a period of field testing and bug fixes before it gets enabled on the Radix mainnet with Babylon.
Q. Can you please publish a blog post sharing more details about Radix Kubernetes chaos engineering?
A. Hahaha - when we get out of the mad rush for mainnet, yea, I am sure we can arrange something!
Q. Why not just do layer two scaling solitons
A. Layer 2s are not really a solution to scaling; they are just a sticking plaster. Today the global financial system does around 2m transactions per second. On DeFi, one transaction can cause 10 - 50 other smart contract interactions resulting from a single transaction. This is because DeFi is highly interconnected vs traditional finance.
This means that if finance moves to DeFi it would need 20m - 100m transactions per second to service the entire world, not to mention the new people we can bring into the financial system by lowering the barrier to entry...
So, no, a layer 2 that can do 7,000 transactions per second is not going to cut it. And quite honestly, the way layer 2 are dealing with state and how difficult they make life for programmers...it's unlikely to be any sort of long-term solution. It MAY become an optimization later, but fundamentally we need to fix the layer 1 scaling issues first.
Q. Will Radix be able to interact with ISO 20022 Data?
A. If you wanted to do that, no reason why not. However, this is very much the old-school concept of data representing the content of another person's database. You could create your own encrypted JSON object and put it on the ledger I guess. It seems like a waste vs. using a smart contract and native financial objects like tokens, though.
Q. How will Radix help in the adoption of real-world assets in DeFi?
A. By making it easier to bring representations of any off ledger asset onto the Radix ledger. We have already started this process with things like Instapass and Instabridge, but we certainly won't have the monopoly on making it happen - any business that wants to do this will have the opportunity to do it successfully on Radix. Things like stable coins are just the start!
Q. I don't know if we know more about this, but can we create logins for users on dApps via the radix wallet on RPN1? or is there some sort of roadmap for this?
A. We'll be releasing more details for things like this around Babylon. Right now we are focusing on the developer tools, the user tools will definitely come, though, and there does need to be a link between both, I agree.
Q. I think the component catalog and developer royalties is a cool idea in theory. However, I'm wondering what protections are given to the developers who create components. Because everything is open source, can't a malicious developer just copy/paste another developer's component and undercut the other developer on royalties?
A. Yep. And there will be people that do that. However, use of blueprints requires trust in the developer that produced them (their reputation) and trust in the people that audit them. Most developer communities recognize the importance of supporting others in the community. It is likely that any developer who did what you are frequently describing would neither be trusted nor supported by the community. These things are more than just simple pieces of code, they are tools for moving, controlling and securing billions of dollars. That seems like reputation and support is going to be essential attributes in the community.
Q. Can you tell us what makes Radix sharding different and unique from other sharding? What does it mean to say your shard space is 2^256? This sounds a little technical. Could you break it down further for the non-tech users?
A. It's a great question, with a very long and complicated answer. The short form is that Radix consensus (Cerberus) is a cross-shard consensus algorithm, meaning that it operates between shards as part of how it functions. All other consensus algorithms operate within the boundaries of a shard. This makes the way Cerberus operates radically different from any other sharding system. To find out more, this blog post is excellent: https://www.radixdlt.com/post/cerberus-infographic-series-chapter-iv
Q. I read that Radix has founded the GoodFi Alliance; what is the purpose of Radix in establishing it? And so far, has the main goal of the GoodFi Alliance been achieved?
A. The Goal of GoodFi is to get 100m people to put $1 into DeFi by 2025. We haven't achieved that yet, but we are in good company (Aave, Chainlink, mStable, Sushiswap, Avalanche etc)
Q. Olympia has been in Beta for two months now, and Betanets are meant to clean things up. So how confident are you guys about Olympia at the moment, and how smoothly is the Olympia working? Also, how useful have the community been so far to make Olympia a success?
A. Betanet has been great. We're closing it down this week and launching Stokenet which is the persistent test network that will run alongside the Radix Olympia mainnet. We have high confidence in both now.
Q. The names and logos of the projects attract attention. Are your logo and name random choices or does it contain messages about your project?
A. Actually, yes, Radix is Latin for "Root", it also a term in number theory meaning "base" - e.g., binary is radix 2 (or base 2). We like to say Radix is the root of the financial system of the future :-D
Q. What do you think is holding the DeFi space back? What is preventing the mass adoption of DeFi products?
A. I think the DeFi space is being held back by three things.
1. DeFi developers find it very difficult to build secure DeFi dApps. There have been $285m in avoidable hacks on Ethereum to date due to solidity. Radix changes this with a purpose-built DeFi programming environment that will enable fast AND secure development. This will let DeFi developers build fast and not break things.
2. DeFi community devs seldom get rewarded for the contributions they make to the DeFi ecosystem. We think this is holding back innovation and new developers entering the space. We are changing that with a system of on-ledger royalties that will reward those that contribute dApp code to the ecosystem.
3. DeFi is super expensive to use because of all of the issues of scalability in DeFi today. We are changing that with our consensus algorithm that will provide unlimited scalability without breaking DeFi composability. This is probably the biggest long-term barrier to mass adoption.
Q. DeFi has evolved from a movement to basically the fastest growing and the most accessible facet of the blockchain industry. In your opinion, do you think it is an opportunity for growth?
A. DeFi is going to eat the world of finance. Ultimately finance is just a big machine for efficiently allocating capital. We the advent of a public ledger that lets capital flow to the best opportunities instantly and that can be programmed to do so automatically, the world of banking is dead, they just don't know it yet. All finance will eventually be DeFi. We are building to make sure that it is Radix that provides that public network where it all happens.
To stay up to date or learn more about Radix DLT please follow the links below.
Join the Community: