 Good morning. Welcome to the Randomest Summit 2023. Thank you all for coming. We've got a day of lots of wonderful speakers and lots of interesting talks. And I will be giving the classic d-rand shill presentation to start off, just so you guys know a little bit about us and why we write the Randomest Summit. So the title of this very brief talk is looking back on three years of d-rand. So in case some of you aren't aware, what is d-rand? It's a threshold network for generating publicly verifiable unbiased and unprotectable randomness. It's built on BLS1231. It uses a Peterson DKG. It was started in 2019 in EPFL, that's the University of Lausanne by Nicola Gaye, who's with us today. It was actually originally a collaboration between Dedes, the decentralized distributed systems lab in EPFL and DFINITY, who make the internet computer. They focus more on getting everything on blockchain and having everything on chain. So Nicola had the idea that maybe it could be generalized a bit to be used in a wider variety of use cases, such as Web 2. So he spun out d-rand. The d-rand mainnet is operated by the League of Entropy, which is a consortium of institutions who run d-rand node. It was founded in 2019 by Cloudflare and some of the other finding partners are here. The goal was to provide randomness free and available as a public service for anyone to use, and also to have a wide variety of institutions running nodes just to keep a little bit of balance. It was then launched in 2020 to support the Filecoin launch. Filecoin is a blockchain, not a dirty word, but it's okay today. You can stay blockchain. We added a whole bunch of members who I won't go through one by one because there are lots of them, but again, keeping a balance of types of institution was one of the key goals here. As of 2023, we've added a bunch of new companies to it. The first one, QRL, run a quantum-resistant ledger, quite cool. Store Swift, who are with us today and will be giving a presentation or a storage provider and do a lot of general Web3 and storage stuff, which is very cool. IPFS Force are a Web3 community that do a bunch of different tooling and open source projects around Web3 stuff. And Automata Network, who are, I think an Oracle service, I remember correctly, and also provide a bunch of different Web3 services. Coming soon, hopefully in the next quarter we'll onboard three more companies. Ken Labs, who again provide Web3 and storage and related services. Dea Data, who are also an Oracle service, and CryptoSat, who put sat lights with cool software into space. And we'll talk more about them a bit later. So everyone loves a good random stat, so I've compiled a bunch of random stats for what we've achieved since version 1 in 2020. First off, we've had 100% uptime over three years, which is magnificent. I think as a testament to the fact that in production, threshold networks can be very resilient. We have 30 billion requests from randomness, at least on our nodes. I haven't got any statistics for Cloudflare, but there's a presentation from Tebow later, so maybe you can give us a little bit of insight there. Including 5 billion requests from a misconfigured Lotus node from the Epic protocol. So thanks Epic protocol for torching our Cloudrun bill. We have 29 releases, and we're currently in version 1.5.4. Next quarter version 2 is going to be coming out, up until now we have no breaking changes. We've been running the same network in production, and yeah, that might change. Additionally, we've had 39 contributors, so thank you to all of those contributors. Hopefully we can onboard some more. Of course, Dea Randa's open source. Yeah, we use the Apache license, so if you want to take the code and do whatever you want with it, you can. And if you want to get involved with code, please find the team and let us know how we can help you get involved. We've also done nine mainnet re-sharing ceremonies of the keys. I think there's not a lot of people running threshold networks in production, and some people question the viability and difficulty in doing some of these things, but we've done it. We've been able to support with two agencies, both well-known, Zondax Ardenlabs, many thanks to them. Unfortunately neither of them here today, but big thanks. So in the last few years, we have delivered a bunch of juicy new features. The first one is support for multiple simultaneous networks. So a network launched in 2020, there was just one chain, and you could only run one chain on a node, but now you can run as many chains as you want on a node. Additionally, you can be a part of different networks with different sets of participants, and maybe at some point in the future, people not affiliated with the D-Round team will run their own network with its own security guarantees. That would be very cool. We also launched a faster network. The initial network emitted randomness every 30 seconds, which coincided with the Filecoin block time, so that was convenient for those guys, but not for a lot of general use cases. We got requests from the community for a faster one in hopes that people could, for instance, use it for gambling apps, which maybe need randomness slightly more often than that. So we launched a three-second network very recently in our main net, and we have dreams of reaching a one-second network at some point in the future, so that we can truly be a cryptographic reference clock. That would be quite cool. Two new schemes. As you may have heard yesterday, if you're out real world crypto, in order to enable time lock encryption, we released schemes that were unchain randomness, i.e. each new random beacon doesn't rely on the previous one that's been emitted. A new one of those schemes also swapped G1 and G2 in order to create smaller signatures for lower costs of storing the beacons or storing the beacons on chain and doing operations with those beacons, which is quite exciting. Nobody else seems to do that in BLS, which makes it hell to integrate with other libraries, but it's a good challenge for us to write lots of new libraries. Also, time lock encryption, which you'll learn presented yesterday at Real World Crypto. Hopefully, you all saw it, and if you didn't, it's already on YouTube, I believe. Those guys have been super, and actually will be presented today as well, in case you missed it, so don't watch it on YouTube. Additionally, multiple storage backends that you can store the beacons in whatever database you would like to store it. Also, I mentioned cryptosap before. They actually brought DRAND to space, so DRAND is running, not part of the Liga Entertainment Network, but is running in space on a satellite, and unfortunately it only has connection with the ground base stations a couple of hours a day, but in Q3, they will be launching a bunch of additional satellites with better coverage, so maybe at some point we will have a full DRAND network in space, but watch this space. Also, we release the types of clients so that people can use it in web UIs or any other JavaScript implementations, and the community has delivered four different Rust client implementations, so obviously, a testament to the fact that everyone loves a bit of Rust these days. So if anyone wants to make another Rust implementation, please do. We also are now available on a bunch of blockchain ecosystems. Protocol Labs released the FEM, the Filecoin Verge Machine in production, I was going to say last month, but I guess it's this month, it just feels like months with all the work put in. So DRAND is in some shape or form available on there. Also, the Noise Network actually launched to Mainnet two days ago. They're building a business on top of DRAND, which is very exciting. It's cool to see companies using our software to do cool stuff, which makes randomness available in the Cosmos ecosystem, which is like a blockchain of blockchain, so as a result, it's available in a whole bunch of stuff. In a hackathon recently, a team called RNGesus made it available on Ethereum until there are still no pre-compiles available for BLS12381, so those guys did a lot of work to build all the necessary cryptography to make it possible to verify randomness beacons on Ethereum, so that's very exciting. Obviously that opens up a lot of new use cases and a lot of new people to use DRAND. Missing Labs also integrated DRAND in SWE, and they wrote some cool demonstration contracts on how to make a lottery and stuff like that, which is very exciting. Also on Aptos, there's work in progress, integrated DRAND there. I don't have any particular news about water space, hopefully there will be some more DRAND on Aptos. So in the next year, we have lots more juicy features to come. We've already released Time Lock Encryption for people who want to do it locally, but soon it will be available in FEM, and I think it will be the first at least built-in Time Lock Encryption on any blockchain ecosystem, so that's exciting. Also asynchronous distributed key generation, one of the challenges of our network at the moment is every time we want to do distributed key generation, everyone has to get to their computer, everyone has to be on a zoom call in case something goes horribly, horribly wrong, which it hasn't yet touch wood. So we're working at the moment to make it asynchronous so that people can run a distributed key generation protocol over a period of a couple of days or a week, just to ease the operational burden. We're also working on better traffic and user analysis. We have a kind of strange reality that we don't really know who's using DRAND. We bumped into Derek Van Willem, one of the Apache Foundation co-finders at a conference in the Netherlands, and he said, oh yeah, I've actually been using DRAND to select COVID patients for clinical trials and I wrote an implementation PHP and we had no way of knowing about this, so hopefully in the future we can actually know who's using DRAND and seek to support them in different ways. We also worked with an offer profit called socialincome.org and attended one of their hackathons. They give out universal basic income in countries in Africa currently in Sierra Leone, so people from I suppose the wealthier parts of the world donate some percentage of their income which then gets dispersed to people in Sierra Leone and they are working to integrate DRAND in order to select these people at random so that some unscrupulous warlord doesn't run off with all the money. So that will be coming probably in the next quarter or two, something in production, which is exciting. As I mentioned before we have a space native DRAND network in the works and that will be very cool. Don't want to say useless but certainly a novelty anyway, it will be kind of cool to have a tamper proof DRAND. Also in the works, thanks to Storeswift who are presenting today they are working on Rust implementation of DRAND. One of our goals to continue the resilience that we've had so far is to make sure no critical bugs can take on the network. So having multiple implementations in different languages hopefully will give us a chance that we have a threshold breaking bug in the network. This is something that's to come, question mark. We have a very old Python client which isn't up to date with all the cool new features. So if anyone in the audience is a particular Python wizard this is a called action that maybe you too could be part of the DRAND experience by developing a Python client for us or updating the existing one. Also we're seeking research students potentially to come up with alternate implementations of the distributed key generation. I know some people in our network and in the related field have come up with some cool schemes recently for doing distributed key generation so it would be awesome in future if we could have a plugable distributed key gen in DRAND and if you know anyone who has any students who are working on such a project we'd love to hear more. Just a random slide for some of the places we've been the last couple of years presenting DRAND. This has mostly been ULAM presenting DRAND on timelock encryption and a whole bunch of conferences including real world crypto yesterday. Where we're going to be in the next year. Obviously we're doing a random assignment right now. The first one in three years so very exciting. The last one was online only. I guess it was over COVID times at that time so now we're back in the healthy world. We can do in person events. Hopefully we'll do some more soon. In two weeks time IPFS thing is happening in Belgium so if anyone is there please reach out to us. We can hack something together. Also the Paris Oracle summit in July. Cowles communication camp in August I think. And very soon we hope to run a hackathon just to get more people building on top of DRAND and hopefully connecting with more people who are building on top of DRAND so stay tuned for news in that direction. And just finally a slide about the team. On the bottom left we've got Yannis who has been covering the project management role for the last 18 months or so. He was only meant to jump in for a month or two to cover and he ended up giving us lots and lots of support so a big thanks to him. He couldn't be here today but thank you Yannis. In the top left is our newest team member Eric who is here with us today. He's the new project lead and joined the team in fact on Monday I think right. So his first week is here in Tokyo where we're crypto and the randomness summit so thank you very much Eric and a massive welcome. In the top middle is Nicola Guy who founded the project in EPFL and is now working on a very cool project in a kind of related space called Medusa which is basically a threshold network for multi-party encryption. Hope I haven't butchered the quick elevator pitch but definitely check it out. Medusa to XYZ I think so it's a very cool project. And then on the right hand side books myself in Yolan or at least ostensibly me. The team who are kind of maintaining and running DRAN at the moment. After Falcone lunch there were quite a lot of people on the DRAN team and it went into the hibernation for a couple of years and early last year it was revived like the Phoenix from the ashes and hopefully in the coming year we can deliver lots of cool new features. So thank you very much for listening. That is all for the last three years of DRAN. Thank you very much for listening.