7/24/24 Discord AMA Summary

Summary of Discord audio AMA, July 27th, 2024.

Space and Time participants:

• Scott Dykstra, CTO and Co-Founder

• Catherine Daly, Head of Marketing

• Mike Post, Community Manager

Thank you to all of the community members who attended! [Session begins]

Mike: Welcome to another Space and Time AMA! As always, we want to thank everyone for attending. And, of course, we want to thank all of you – the community – whether you’re new here or you’ve been with us for awhile. Man, it’s been almost two years for some of you now! We’re always so appreciative for all of your support, all of your feedback, and we’re continuing to look forward to big things as we move through 2024. I’m here today with Scott and Cat. Cat, for those here who don’t know you, why don’t you tell us a bit about yourself?

Cat: Hey, everyone. GM, thanks for coming to the AMA. We're super excited to have you here. I see we have a lot of new faces! So if this is your first time, let us know in the chat. We're super excited to have you here. I'm Cat. I'm the Head of Marketing at Space and Time. I’ve been with Space and Time for about two years now, so I know a lot of you very well, and I’m looking forward to getting to know the rest of you!

Mike: Scott, it’s your turn!

Scott: Absolutely. Yeah, thanks to everyone for showing up. It's great to see everyone here, so many familiar faces and new faces in the Discord. And by faces, I mean anonymous profile pics. Love you guys! I'm Scott Dykstra, Co-Founder and CTO of Space and Time. I’ve been working on the Space and Time project since about summer of 2021, so about three years now, and it's been a really magnificent journey in building systems that can power the next generation of smart contracts and, more recently, the next generation of AI agents that transact onchain. We're gonna have a fun Discord AMA today,

Mike: So let's dive right into the questions here. We're going to start with some questions we got about the product and the ecosystem. And as always, thank you to all of you who submitted questions! So this one's for Scott. We launched Proof of SQL recently. Can you tell us more about it? And what is the difference between Proof of SQL and other ZK co-processors?

Scott: Yeah. Okay, two very good questions. So, Proof of SQL is a way for smart contracts to ask questions about activity on their own chain or other chains. And I, you know, come from a database background. I built enterprise databases for the Fortune 500 for about a decade, and as we got into crypto and smart contracts during COVID in 2020, we quickly realized there's a huge need for a way for smart contracts to be able to access data, and particularly historical data, from their own chain, in order for smart contract developers in the DeFi space to be able to build more powerful, more sophisticated, more sophisticated DeFi protocols. After we launched Proof of SQL, I was kind of surprised by the amount of inbound requests we got from cool projects in the crypto space that had their own ideas and their own use cases for how they wanted to use Proof of SQL to power their protocols in ways that I didn’t anticipate. For example, we talked with some restaking protocols like Karak that wanted to use Space and Time as their slashing mechanism for slashing restaked nodes, and they realized pretty quickly, Hey, we can just run SQL queries against the Space and Time network to get back information about whether or not a certain a certain node has been malicious or done what it’s supposed to do or complete a job within a certain time limit. And using very simple standard SQL, we can just have a smart contract ask questions about, Hey, show me all nodes that didn’t complete a job within a minute, and just slash them, which completely opened up a whole new space for DeFi developers.

The second part of the question is how Space and Time is different from other ZK co-processors and other zkEVMs. Well, as far as I know, most, if not all, ZK co-processors in the space are focused on generic, arbitrary compute. The problem with that is that your performance is very slow. We’re talking minute proof times, sometimes five, ten, twenty, thirty-minute proof times – which is fine unless you want to actually facilitate transactions. For example, if you’re a user onchain and you’re going to take out a loan onchain against Ethereum, when you want to take out that loan, then you want to wait maybe a block or two before some sort of confirmation gets back to you. You don’t want to wait five minutes. So while most ZK co-processors focus on arbitrary compute – like, letting a DeFi developer write any kind of calculation – they’re really slow. What we did is take a different approach. We said, Hey, what are the fundamental things that DeFi developers need? They just need to access data. They can already write calculations in their smart contract. They need to do DeFi calculations that can be done in a smart contract or in a different zkEVM or in a different co-processor, but the biggest burning need that we know DeFi developers have is just accessing data. Like, Hey, before Scott’s wallet takes out a loan, let’s go visit Scott’s wallet’s history onchain, and see if Scott has paid out some loans onchain in the past, and if so, then let’s give Scott a better borrow rate, let’s give Scott a discounted loan. Or Hey, before letting Scott LP into this liquidity pool, let’s look up his wallet history and see if he’s staked one of the tokens in that pool for at least six months, and we’ll let him in. Or Hey, before Scott gets this reward for being a gold-tier video game player, let’s make sure that he’s minted our NFT, if he has at least 10,000 points, and if he has at least three hours of gameplay, and let’s provide that information to a smart contract just in time during a transaction to give the smart contract the data that it needs to make decisions.

Finally, things evolve. Like, over the last three years, AI became very important to us and to the whole world in ways that nobody anticipated three years ago. And so we saw a huge opportunity to also provide that low-latency verified data to AI agents that are transacting onchain.

Mike: Scott, could you tell us about some of the projects that have already implemented Proof of SQL, about the use cases they’ve applied it to, and – just to add on to that – are there any Proof of SQL project use cases you’d like to highlight?

Scott: Yeah, I already highlighted two. One was a lending protocol use case. The future of lending is the lender looks up your wallet information, queries your wallet history just in time during a transaction to see if you can kind of build a credit score, build a profile around your wallet, and maybe give you discounts on your borrow rate. We're supporting a couple lending protocols around that kind of onchain credit score, supporting restaking protocols around how you slash node operators in your network when they don't do things they're supposed to. The challenge with restaking in general is that most restaking node deployers don't write slashing logic. There's no slashing that's occurring at all. So it's kind of a meme. Like, hey, we're restaking our capital against these different protocols to secure them with our capital. But there's no slashing logic deployed today, so what's the point? Space and Time is helping with that. We're making it easier for DeFi developers within these restaking protocols to write slashing logic very quickly. We are the security for those protocols. We're working on deploying a new framework called zkETH, from the Space and Time team, that's launching probably about a month. Cat and I are working through the exact date we want to launch this, but zkETH will be a way for our ZK co-processor to power a single, unified account for community members to kind of unify the balances across all their wallets across a lot of different popular EVM chains. And there's more to come on that. It’s a really exciting framework powered by our ZK co-processor. And then a couple other customer use cases. We're helping a few games and a few loyalty rewards projects. There's a couple of really interesting teams building sort of a Zealy for on chain activities, and we're helping them ask the questions they need about which wallet actually completed the onchain quest. What DeFi developers need, as I'll continue to say, is really just a way to access data that they need to make decisions onchain about how to reward people.

Mike: Thanks. Scott. It's been very exciting seeing all of the novel use cases of Proof of SQL. It's really going to make a big change in the space overall. So, moving on, we announced a big partnership back in March with ZKsync, and part of that was an elastic chain we're going to develop on ZKsync. For those who are less technically knowledgeable, could you explain the ZKsync elastic chain in simple terms, and also, is it comparable to Polkadot parachains or the Polygon AggLayer?

Scott: Good questions. Wow. You guys rock. There's so many good questions you're hitting me with right off the bat. I'll start with the end of that question. First, is it comparable to Polkadot parachains or Polygon Agglayer in a way? I think the ZKsync team, the Matter Labs team, took a lot of inspiration from the design of the Polkadot parachain ecosystem. And I think the Agglayer approache is really, really wise as well. I think what Polygon is doing is very innovative. The difference is that this is a bunch of chains that can not only roll up to Ethereum, but can also kind of interop with ZKsync Era, their main chain. But, yeah, it's analogous to a Polkadot parachain in the sense that it's, you know, a set of chains, a set of L2s that can interop with ZKsync’s main L2, Era, and roll up to Eth. Now, the reason we chose to build our own L2 on the ZK chain stack is that we’re very aligned with ZKsync that ZK is the end game and also, what you're going to find out soon, if not already, is that more is about to be revealed to the crypto community that the Ethereum ecosystem is diverging a lot and moving more and more towards a ZKsync-native style of rolling up. ZKsync made a lot of trade offs that were good trade offs. They removed a few opcodes, they added support for their own opcodes and for non-technical folks. All that means is ZKsync has removed some of the Ethereum functions, like some of the Ethereum capabilities of the EVM, and added a few of their own. And Vitalik likes that. He's been blogging about how that's actually the right way to go for better performance, and I just saw this morning, that the Ethereum roadmap is going to start to kind of go that direction as well, and ZKsync already supports it, so while everyone else has to play catch up, ZKsync is a little ahead of the game. And the Matter Labs team is hardcore right there. They are hardcore engineers building a very high quality zkEVM, and we just wanted to be on a high quality, very low latency prover aligned to our thesis that ZK is the end game. Now, I think very highly of Polygon CDK as well. And I think there's going to be a lot of tough competition between Polygon and ZKsync, because they're both very good options. They're both very good zkEVM deployments, but ZKsync is a little ahead of the game in terms of making tradeoffs in terms of opcode supported for better performance. And then the analogy of the Polkadot parachain is an interesting one. I think it's a really good analogy that, you know, instead of a bunch of, essentially layer one blockchains with 15 to 50 nodes, it's just a bunch of L2 provers that are all interoperable. All right.

We haven't decided on a definite date yet. We’re still working on that. That will be a huge announcement. Now, I know this sounds crazy, but we're also building an L1 on top of substrate to power all of our data operations. Our architecture is very unique. We basically have a ZK co-processor L1 with 30 to 50 permissionless nodes and growing that we're launching later in the year that is essentially an L1 with BFT consensus to kind of gather all the blockchain data that our network has collected and prepare it to load into the ZK co-processor and sort of do all the data prep. It's an indexing chain. It's a database chain that indexes blockchain data into a database that the prover can read to. Now, the reason we primarily focus on our ZKsync deployment is that's where all of our liquidity is. In the same way that Avalanche has a C-Chain and an X-Chain, we will have a database chain that's built on top of the Polkadot stack. Ironically, given the previous question, and an L2 from ZKsync, where all the liquidity EVM contracts and apps are deployed that use Space and Time’s ZK co-processor. So basically you have this large L1 or permissionless nodes where anybody can join and bring their own prover and access all the data and that handles all the data operations, running queries, building proofs, prepping the data we've indexed from other chains, and you have an L2 where all the EVM contracts are deployed that call the ZK co-processor and ask questions. So if you're building a DeFi protocol you can deploy on our L2 but your DeFi protocol can get data that it needs from an L1, I know that sounds like a very crazy architecture, but you're going to realize in like, two years, everyone's going to start copying Space and Time and doing the exact same thing, and you'll see why we always tend to be a year ahead of the curve. So maybe in a year, there'll be, like, eight other teams that are like, Hey, we're deploying an l2 for our liquidity and our EVM apps, and we're deploying an L1 via the substrate or Tendermint Cosmos ecosystem to do more technical data operations, to prep data, sign it, load it into a ZK prover and make it available to the ecosystem. I'll be publishing a whitepaper on what the heck this all means, but the primary focus for Space and Time right now is just getting our ZK-chain launched and connected into the ZKsync ecosystem.

Mike: That sounds great. Things just keep getting bigger and bigger, and you and the rest of the engineering team, wow, the sheer scope of your innovation, technical skill just continues to floor me. It's really been amazing to watch the scope of Space and Time just continue to grow over the last couple years. So let's talk about another aspect, more of a community facing aspect, actually, of the platform, which is the Space and Time Studio. And we've been thrilled to see all the dashboards that all of you have been creating. And a special shout out, actually, to Closebox, who in the Aptos contest just took it to the next level and created the best dashboard we've ever seen. So at the moment, the only way for users to pay for compute credits on Space and Time Studio is a credit card. And I know a lot of people would like to be able to pay for compute using cryptocurrency instead. Are there any plans to make this an option for the future?

Scott: Yeah. So I mentioned a minute ago this framework that we're launching called zkETH is essentially a cross-chain unified account balance across all your different EVM wallets on popular EVM chains. And what we're doing with that is we're allowing you to use that account balance to pay for your Space and Time compute onchain. That will be a way that the community can just pay for queries onchain and procure some Space and Time compute onchain. So basically, you can deposit any popular ERC-20, like USDC, USDT, LINK, ZK, you know, popular hard money and some tokens from the ecosystems we're closest with, of course, and that gets added to your unified account balance. That kind of unifies all your wallets together across chains. And if you want, you can optionally spend some of that balance onchain to pay for queries. More to come. But we'll be launching onchain payments along with the zkETH unified account balance framework in about a month, and we're going to try to make the make this as exciting as the Proof of SQL launch, because this is something we've been working on for a long time, and it's very technical, but it's for end users. It's for consumers, it's for retail participants. It's not for technical people. It's hard work; we did very tough engineering. But it offers something that everyone can use. It's like Venmo for crypto, a way to send gasless transfers to other accounts without paying gas, other zkETH accounts, including Space and Time's own zkETH account to pay your bill.

Mike: Let's talk about one of our favorite subjects, which is AI. Space and Time has partnered with Microsoft to allow developers to validate their data for AI training. Can you elaborate on any other AI-based solutions that Space and Time plans to implement in the near future?

Scott: Yeah, I don't want to say too much, because I don't want to be the guy that's the whole AMA. I'm just like, hey, coming soon. We're launching this coming soon. Coming soon. Stay tuned. Stay tuned. But we're co-engineering a relayer for GPT-5. We're co-engineering that with Microsoft right now. Very hush hush. We're not saying much about it. We kind of just want to launch it out of nowhere. We kind of want it to be a surprise. So, everyone, all 324 of you, this is between us! This is just Space and Time, fam. You heard it here first: we've been co-engineering, with Microsoft, a relayer to bring GPT-5 onchain once it's ready. Rumors are swelling. Rumors are circulating that GPT-5 will be released a little later this year, likely in early Q4, and so we've been prepping for that release by co-engineering, with Microsoft, a relayer to access GPT-5 from EVM contracts. So it's a way for smart contracts just to directly call LLMs, particularly very sophisticated ones, and it's secured by an enclave, a Trusted Execution Environment, a TEE, if you're familiar with those. This is a project that we've been working on for a while. So it's kind of outside the scope of what we typically do. Most of our research is all about very, very low latency ZK-proofs to provide data to smart contracts. But with that, hey, why don't we also provide LLM inference? And let's do it from the next generation of LLMs. Let's just build for GPT-5. So, more to come. That'll be a big announcement that we’ll announce with Microsoft. Whenever GPT-5 is available. Who knows when that’ll be? It could be next year, for all we know, but I'm very hopeful that it'll be around September. It's really hard to know. OpenAI is very hush-hush. They don't tell us much.

Mike: I'm looking forward to that as well! Hopefully GPT-5 will be released sooner rather than later. So let's move on to some community questions for Cat, and then we'll visit roadmap questions after that. So, Cat, we've been running the advocate program for a little while now, and we selected our first cohort back in May. When are we planning on adding more advocates?

Cat: Yeah, great question. First off, I want to shout out our current advocates, because you guys have been so awesome over the last few months. I can't tell you how much we appreciate everything that you do. We have ten of you right now. We do a monthly call together, and you guys are truly our best ambassadors. We obviously selected the ten of you from the people who have been engaging most closely for the last couple of years. So you've always been awesome, but you guys have really gone above and beyond as our advocates. So we love having you on board, and we're looking forward to adding more people to the mix. We don't have a concrete plan or timeline for how many advocates were onboarding and when, because we'd rather bring people in as there's need as the community grows, and as we see more people who are really going above and beyond in the community. So I'm excited to share that we'll be adding a few new advocates to the program over the next couple of months, so keep an eye out for that. And we'll probably plan to reevaluate about once per quarter, but again, it's really just going to be based on growth and need. We're really excited to keep that program expanding.

Mike: Absolutely. It's been great. It's been a blast watching all the content the advocates have created and so some wonderful working with you guys and hearing all your feedback. So are there any additional plans beyond just the advocate program for promotion and marketing as we near testnet and beyond? And how can community members better engage as the project builds toward that phase?

Cat: Yeah, there are some really exciting milestones up ahead for Space and Time! Scott's mentioned a few of them. We'll talk about more of them in the roadmap section. But I think the most important thing that we as a community can do right now and over the next few months is really spread the word about Space and Time and get the buzz going. So that's how I encourage all of you to engage. And many of you have already done an awesome job at it, some of you for, like, almost two years now. So keep tweeting about Space and Time. Keep making infographics, keep making memes.We love to see those, and they get a lot of traction in crypto Twitter. So we definitely encourage you to do that.

I would also encourage you to engage with everything Space and Time posts on Twitter – you can actually turn on post notifications so that you're notified when Space and Time tweets, and then go like it, retweet it, reply to it. The more you do that, the more visibility our posts get, and the more buzz is generated. Invite your friends to the community, share the stuff that you're doing, like your DIY projects – those have been awesome – in other communities that you're a part of. I run marketing for Space and Time, but truly, you guys run marketing for Space and Time. You guys are our best ambassadors. You guys are our best promotion. So let's get the buzz going, and let's keep the community growing as quickly as it has.

Mike: For sure. And speaking of memes and content, everybody's been wanting a community mascot for some time now. Do we have any updates on that front?

Cat: Yes! We have put a ton of thought into this, and I have been cooking this up for weeks now, and I'm really excited to introduce our community mascot to you. So if you head over to the Space and Time Twitter right now, you'll see Gigabrain, the official Space and Time community mascot. You know, we figured our team is made up of Gigabrains. Our community is made up of Gigabrains. So who better to represent Space and Time than a gigabrain? So we want to see Gigabrains everywhere. You guys have been making awesome memes with Pepe, the Simpsons, and other characters, and we want to see them with Gigabrain as well! So Mike is going to put up a new channel in the next few minutes that has different versions of Gigabrain doing different poses, which you can use to make memes. If you're not sure how to do that, it’s super easy – go to Midjourney or Dalle, generate a background, pop your favorite Gigabrain image in there and tweet it out. We are going to be randomly selecting ten of you who attended this AMA to receive a V5 Space and Time Community NFT, and we're also going to be randomly selecting ten of you who engage with our Gigabrain tweet to receive a V5 NFT. So here's how you enter: go like the Gigabrain tweet, retweet or quote tweet it, and reply below the tweet with gmgb, which stands for “good morning Gigabrains.” And if you do all that in the next 24 hours, you might win an NFT.

Mike: I've been looking forward to this mascot for a long time, and it definitely turned out great. I like it. So finally, on the community side of things, are there any updates on Zealy in terms of where we are and where it's headed?

Cat: Yeah. So for those of you who are new to the community, I want to stress that we've always tried to make it clear that Zealy is a fun way to measure and reward community contribution. There's no real-world value tied to points, no promises of what those points translate into. But as an example, your community points get you access to certain benefits within the community, such as the opportunity to apply for the advocate program, or to be eligible for certain community contests that we run, and possibly more rewards in the future. It's mostly meant, though, to help guide you all on how to engage with Space and Time. Like I said, we love to see your memes, your blogs, your videos, but honestly, one of the best ways that you can engage is by actually using Space and Time. As Scott mentioned, it's super easy to sign up on the Space and Time Studio, connect your wallet, and sign up for a Pay-per-Compute subscription. Today we support credit cards, but we’ll soon be supporting onchain payments. So start running queries, start building dashboards. If you're looking for new ways to engage, or ways to engage better, or alpha and what we want to see more of, I would say, just use Space and Time.

Mike: Great to know! So let's move now back to Scott for some roadmap questions. Scott, could you update us on what is next in 2024? When are testnet and mainnet planned for? And are there any other key milestones coming up?

Scott: Yeah, testnet. I don't want to give it the exact date, as that's still being worked out with the ZKsync fam, and we're waiting on a few-last minute code releases from Matter Labs to help us out get ready for our testnet. Cat doesn't like me speaking to specific dates, because sometimes they change.

Mike: That’s a huge step!

Scott: Yeah! I'm gonna go to sleep, wake up, and it's gonna be here. All this hard work we've been doing over the last three years is going to finally come to fruition.

Mike: Yeah, it's great to see. I'm really looking forward to that. So how soon after testnet begins will users be able to operate nodes? And will it be only big node operators during testnet, or will individuals be able to begin opening them as well?

Scott: At mainnet, our L1 will be permissionless. Remember that earlier in the conversation I talked about how we're deploying our L2 rollup with the ZKsync stack. We're also deploying an L1, a sovereign chain, for all the database operations, and that'll be permissionless. Anybody can run a node, just take some Space and Time token and participate in the network. I just leaked some alpha in saying that!

Mike: Everybody likes alpha! So how is testnet going to function, and how will regular non-developer users be able to participate?

Scott: I promised that I wouldn't be the guy saying coming soon, coming soon. But yeah, we’re building a little AI studio to make it easy for non technical folks, people that don't write code be able to build little mini apps and kind of generate them with AI. My fun little smart contracts and front end applications, like you could, like, generate a little sports betting app on Space and Time chain with your friends. Or you could generate a little prediction market for the election and deploy it yourself and add liquidity. And you could build a little app for, you know, yield farming, to follow a strategy that you made, and so we're just making it really, really easy for non technical folks in the community to not just build dashboards, but now also to build little applications. It's kind of the next evolution of Space and Time. It's something we've been working on for a while now. I’m hoping we'll be ready by the time we mainnet.

Mike: “Coming soon” or not, all of this is super exciting. Like you said, all of the hard work is coming to fruition.

Scott: We're shipping a lot right now. We've been talking about Proof of SQL. We shipped it. We're shipping a unified account balance framework on top of Proof of SQL, we’re launching testnet. Then, later on, we are launching our little AI studio and a relayer for GPT-5 to get LLMs onchain and build little onchain applications on our chain. No promises, but we’re hoping to get all of this out in 2024!

Mike: All right, let's move on to our final question here, and I'm going to dish this one to Cat. As always, one of the most common questions we get is “when token?” What can you tell us about that?

Cat: So first I want to say that it's been awesome to watch how quickly our community has grown. We're about to pass 200,000 followers on Twitter, and we’re seeing a bunch of new faces in Discord all the time. The DIY contest that we're currently running is by far the best we've ever seen. So we're really grateful to all of you for your continued support and engagement with the Space and Time community. We're all excited about a potential Space and Time token. I know you all are, and we are too. It's really important to us that we do it the right way. And Scott highlighted some key milestones that we're working toward aggressively right now. We want to get testnet up and running soon, and mainnet after that, and throughout those milestones, we're also going to make sure we're moving toward decentralization, true decentralization. So that's a key focus for us.

It's important as we mature and decentralize our network that the network participants are incentivized and rewarded for their contributions, and that those of you who have been part of the community, especially those of you who are engaging with the Space and Time Studio, running queries, building dashboards, get to share in the collective success of Space and Time. So as always, keep an eye on our announcements. Keep engaging, and I think I can say confidently that it's going to be a very exciting rest of the year for Space and Time.

Mike: It’s so exciting! So that's going to be it for our questions. So, Cat and Scott, do you have any closing remarks before we finish for the day?

Scott: I see a lot of folks mentioning sybils here on Discord. I'll just quickly say that you shouldn’t worry too much about sybils. We're going to go way, way out of our way to kind of prevent sybil attacks and block sybils. I've been building some sophisticated infrastructure to prevent sybils. But more importantly, I’m just really, really, really proud of this community. It's grown so much in the last year, especially in the last six months. Cat and Mike have been working their butts off, but also it's just really encouraging and very heartwarming to see all the support. We don't have the biggest community, but we certainly have the most heartwarming and the closest community. You guys really rock, and we're going to return the favor. Our team's working really, really hard to ship this year and give you guys something that rewards all your incredibly hard work. We really appreciate it.

Cat: Yeah, I want to echo what Scott said about engagement. Obviously we have rules for how you can engage on Zealy. If we ever see anyone replicating content across two different accounts or copying someone's content, that's an automatic ban. Beyond that, though, we have plans for sybil prevention before there are any rewards associated with Zeely points, so don't worry about that too much. We'll take care of it. And also to echo what Scott said, you guys have been so awesome! Seeing the community grow the way it has, it’s been absolutely just mind blowing. We never expected to see it grow as fast as it has, and we're super excited to watch it continue throughout the rest of the year as we approach some exciting new milestones.

Mike: For sure. Like I said at the start, your support and feedback really just means everything to us. And I love my job., because I'm the one who gets to interact with you guys on a regular basis. So as I've always said, anytime you have feedback, questions, or concerns of any kind, please get in touch with me. And to echo Cat and Scott, we're really looking forward to having all of you with us as we move on to the next steps! So thank you, all of you, for attending. We'll look forward to seeing you next time.

Cat: Thanks, guys! [Session ends] Space and Time Links:

Website: http://spaceandtime.io

Twitter: http://twitter.com/spaceandtimeDB

Discord: http://discord.com/spaceandtimeDB

Telegram: http://t.me/spaceandtimeDB

LinkedIn: http://linkedin.com/space-and-time-db