Austin Federa: From Solana Foundation to Double Zero's Fiber Revolution

Join Alex Golding as he sits down with Austin Federa, Co-founder of DoubleZero, to explore how they're building permissionless high-performance fiber infrastructure that could revolutionize blockchain performance. Austin shares the technical vision behind creating a parallel internet for distributed systems, starting with Solana validators as their initial market.

DoubleZero: https://doublezero.xyz

🎯 Key Highlights

▸ The internet bottleneck: why high-performance blockchains are hitting infrastructure limits

▸ Building a parallel internet with dedicated fiber, FPGAs, and multicast technology

▸ Proof of utility: weaponizing greed and capitalism for positive-sum network outcomes

▸ 16% of Solana network already running on testnet despite limited infrastructure

▸ Technical deep-dive: signature verification, packet deduplication, and jitter reduction

▸ Token economics and bandwidth contribution models vs traditional staking

▸ Long-term vision: replacing IP addresses with public-private key routing

💡 Subscribe for more crypto & infrastructure insights! 🔔

🧠 Follow the Alpha

▸ Austin's Twitter: @Austin_Federa

▸ DoubleZero's Twitter: @doublezero

🔗 Connect with Delphi

🌐 Portal: https://delphidigital.io/

🐦 Twitter: https://x.com/delphi_digital

💼 LinkedIn: https://www.linkedin.com/company/delphi-digital/

🎧 Listen on

Spotify: https://open.spotify.com/show/62PR1RigLG2YN5Pelq6UY9?si=18ac7ccf36ab4753&nd=1&dlsi=50105fd66e6c4124

Apple Podcasts: https://podcasts.apple.com/us/podcast/the-delphi-podcast/id1438148082

Youtube: https://www.youtube.com/channel/UC9Yy99ZlQIX9-PdG_xHj43Q

Timestamps

00:00 — Intro: Austin Federa and the internet bottleneck problem

02:00 — Building a parallel internet for distributed systems

06:00 — The Firedancer project and million TPS potential

09:15 — Technical architecture: fiber, FPGAs, and multicast technology

13:30 — Signature verification and packet deduplication explained

17:15 — Proof of utility: Shapley values and economic incentives

23:45 — Validator onboarding and network connection process

29:30 — Testnet success: 16% of Solana network adoption

33:15 — Latency, jitter, and bandwidth improvements explained

39:00 — Token economics: 2Z token and payment models

44:45 — Access models vs staking requirements for enterprises

51:30 — Validators.app badges and network transparency

59:15 — Governance roadmap: from mainnet beta to full decentralization

01:05:45 — Long-term vision: replacing IP addresses and DNS

01:09:15 — Lightning round: most misunderstood aspects

01:12:30 — Five-year vision: parallel internet for high performance systems

Disclaimer

This podcast is strictly informational and educational and is not investment advice or a solicitation to buy or sell any tokens or securities or to make any financial decisions. Do not trade or invest in any project, tokens, or securities based upon this podcast episode. The host and members at Delphi Ventures may personally own tokens or art that are mentioned on the podcast.

Transcript

SPEAKER_01 0:01

You're now plugged into the Delphi podcast.

SPEAKER_00 0:06

Hey guys, uh Solana and other high throughput chains are hitting internet limits, not software limits. And for that reason, I'm here today with Austin Federa, co-founder and CEO of Double Zero, about building a permissionless, high performance fiber pipeline, starting with the Solana network and its validator set as an initial target market. We're gonna dig into how it works, what changes for operators, and how to keep it fair. I do need to disclose Delphi is an investor in Double Zero, and nothing here is investment advice. I also need to disclose that I'm super bullish about Austin and this project, and for months I tried to gain access to it, and even uh you know wrote him a message from my hospital bed being like, Austin, I just can't stop thinking about your company. Uh, we have to, you know, somehow work together on this. So, with that all said, um, I'd love to turn it over to Austin so he could do a brief introduction and we can kind of take it away from there.

SPEAKER_01 0:59

See, now I think your interest in the project was just the pain meds. Yay, thanks for having me here today.

SPEAKER_00 1:08

Of course. Um, so as you guys know, uh, you know, I'm Alex Golding, I'm a VC here at Delphi Ventures. Uh, before that, I worked at Solana Labs briefly uh in the summer of 2021 on our business development and uh venture capital teams. And while I was there, I met Austin and I got to know Austin, I got to experience his work ethic, his per his personal uh got to watch the way he comported himself, the way he did business, the way he interacted with others. And um now, you know, we get to come along for the ride with double zero. Um, as I mentioned, you know, they're solving, they're working on a problem that's solving uh some something that affects the entire internet. And, you know, I'd love if maybe he could introduce himself, uh, give a give us a brief background and what brought him here, what inspired him, and yeah, we could take it away from that.

SPEAKER_01 1:52

Um so I'm one of the three co-founders of Double Zero, and we really are building a parallel internet for high-performance distributed systems. And effectively what this is is we've been limited in not just crypto, but mostly crypto for a number of years by the fact that everything relies on the public internet. And that may seem a little bit like an oxymoron because you look at these companies like Google, Amazon, Microsoft, Apple, Facebook, you think, oh, these are all giant public internet companies. Like, how is the public internet holding back these organizations? And the answer is it's not because they've built entirely parallel networks. We think about these as public internet companies, but Meta is the largest investor in undersea fiber of any company in the world right now. Uh, and they do this not out of the goodness of their heart because they want to make the internet a better place. They do this because they want to sell you ads and they want Instagram reels to load really quickly when you're in Malaysia and they want to be able to have video calls between people who are in different parts of the world. They want to basically stick you into their ecosystem so they can sell you ads. Most of these companies are ad tech companies at the end of the day. And in order to do that, the public internet is too slow for them. They've built this whole parallel infrastructure. But private networking is an anti-network effect business. If I build a private network for my company, your company can't use it. And this is kind of how it's always been built. The first folks to really build private networks were, well, after the universities who had built the public internet to begin with, um, were financial companies that are going out and trying to build systems like high-frequency trading systems. But we've never had a private network, a dedicated high-performance network that was operated by multiple independent parties. And if you think about it from a crypto perspective, no one's going to trust a network run by one company. It needs to be run by a bunch of different companies and a bunch of different individuals. And so that's really what we set out to build with Double Zero, is a high-performance network that can have the same performance characteristics as Google and Microsoft and Apple and Amazon and all these big tech companies use, as well as the high-frequency trading firms like Citadel and Jump and DRW, but is owned and managed by multiple independent entities and is therefore suitable for blockchains and allowing them to reach their true potential.

SPEAKER_00 4:22

And so what inspired you to take this on? Like what you know, what gave you the idea that this was even an opportunity? I mean, you had this great role at the Solana Foundation. And so after you saw this opportunity, what convinced you to then leave Solana and go found this?

SPEAKER_01 4:37

Yeah, so I joined Solana in late 2020 after working for Bison Trails, uh, which got acquired by Coinbase and became Coinbase Cloud. And so in the four years at Solana, I sort of you know sat there and saw the rise of the protocol from, you know, I think a dollar 25 when I joined all the way up to the uh you know the highs of 2021, then the lows of post-FTX uh back to the place that it is now today. And through that process, you really see like engineering getting much more performant, uh, really heavily optimizing the validator client. One of the projects that I had some involvement in here was the Fire Dancer project, which was an effort to build a new validator client for Solana that could do a million transactions per second, a number that sounds comically high if you go back to 2022 and look at what these systems were doing at the time. But now on Solana, we have two clients that can push a million transactions per second. We have the Agave client built by the Anza team, and we have the Fire Dancer client built by Jump Crypto. However, neither one of them can reach their true potential on the public internet. And this was kind of something I started hearing from some of the engineers in late 2022, was like, hey, we're we're building this incredibly performant client, but it's not gonna be able to hit these speeds on the internet. And in my head, I was like, okay, well, why is this? Why are we spending all this time and effort building a client that we're not going to be able to reach the full potential of? And they're kind of like, well, you know, it's it's kind of like you want a car that can go 200 miles an hour, even if you can never actually drive it over 90. And it was like, okay, I sort of get that reasoning, but like it feels like we're leaving a lot on the table here. And so this idea of could we build something that blockchain people would not trust, but be able to verify and have the sort of uh censorship resistance assurances that you want if you're gonna operate a network on something like this that would enable this validator client and others to reach these types of performance metrics. At the same time, you know, you'd see like Apdos and Suey publish these benchmarks that show like 700,000 transactions per second in a testing environment. And you see them launch and they're doing several thousand transactions per second, like totally respectable numbers, but nowhere near what their sort of benchmarks said they should be able to do. And you dig into that, and you know, some of its demand, some of its program optimization, a lot of it's the public internet. And so I started to see that there was this problem that was not Solana, like the problem Solana has are usually two to three years ahead of where the rest of the industry is. And so if we say, okay, Solana's facing the limit, the fundamental limit is the public internet now. Two or three years from now, I expect most major high-performance networks are going to be limited by the public internet. So could we actually build something that can address the problem for Solana, can unblock a bunch of this for developers going forwards and help new upcoming blockchains be able to reach hundreds of thousands of transactions per second as well? And then also applied in other areas like storage and CDN and gaming and AI workloads. Um, and that was really what we set out to build with Double Zero. And, you know, one of the kind of funny things is it's like, oh, like what do these crypto people know about building networks? And the answer is not a ton. This the, you know, the only reason that this project is possible, and then we've gone from announcing to getting ready to launch mainnet in less than a year, is because of the incredible engineering team and the two co-founders on the project, Mateo Ward and Andrew McConnell, who spent their entire careers working in high performance private networking and dedicated networking. Um, and so we've been able to sort of bring together the best of what traditional finance has to offer in terms of high performance dedicated networking that's competing on the microsecond and nanosecond level to the crypto space where no one really thinks or knows much about networking. If you really look at the average level of core protocol developer, it's sort of still seen as this like, well, we just have the public internet, we don't have to optimize it too much. Or if we're optimizing it, we're optimizing it on top of the public internet. And so I'm really excited to sort of bring this new toolkit to protocol developers and sort of see what they come up with and what limits they didn't think it was possible to push that they now can, because all their validators have 100 gigabits of connectivity between them.

SPEAKER_00 9:01

That's awesome. And I got a lot of follow-up questions, but unfortunately, I can't ask one at a time. Um, so before we dive too deep into things, could we maybe level set and just help people understand uh in one sentence, uh, you know, what the connection is between Meta's fiber and double zero's vision? Um, you know, are you guys gonna be participating in Meta's fiber? Are you gonna have your own fiber? Do you have anything around, you know, any type of tech around that or around your fiber lines? Um yeah, can we just please?

SPEAKER_01 9:33

So similar to like on a on a compute-based blockchain network like Solana or Ethereum, the Solana Foundation doesn't run the validators, the Ethereum Foundation doesn't run the validators. We don't actually run fiber at double zero ourselves. What we what we build and manage is a software orchestration system that runs on basically the switches, similar-ish to the switches you might have in your house and the router you might have in your house, just much beefier and much more expensive. Um, and that manages the mesh process of building out this high performance network. And so we don't have our own fiber. We're working with contributors that sometimes have their own fiber, sometimes they've leased it. Um but this is typically how it's done in the industry. When Facebook is building fiber, it's not like Facebook owns a bunch of fiber building ships. They're basically providing all the financing and then they own the output product for it. Um, and so one of the cool things about what we're doing is we can work with all different types of fiber connections or other types of connectivity as well. And so you can have, you know, a fiber system where some of it's contributed by, you know, we have we have 11 contributors today contributing fiber to the double zero network. Some of them are trading firms, some of them are bare metal providers like TerraSwitch and Latitude and Cherry Servers. Some of them are venture capital firms that have decided, like, hey, we actually are gonna both invest in double zero and we're going to take fiber and take leases on it and contribute it to the network. The same way that, you know, if our if someone were building a DeFi protocol, their investor may deploy a bunch of capital onto that DeFi protocol or something along those lines. Um, in this case, the way you contribute to the network is not by running a validator, it's by providing fiber connectivity into it.

SPEAKER_00 11:21

Oh, very interesting. Um so we're I'm gonna put a pin on that for later because I do want to ask how the community can get involved. But first, let's just continue with the level setting on the technical side. So um you're onboarding fiber, sounds like a D-Pin network, right? Um, and I remember you know that you guys have uh deduplication uh uh software. Yes, you've got also a hardware component with FPGAs, um, you've got this like inner-outering architecture. Could you just you know explain it uh for everyone, please?

SPEAKER_01 11:52

Yeah, so I'll say to for sort of like product category, I like D-Pin projects. I think they're quite interesting. I think what Helium has done and been able to do is really interesting. One of the challenges with calling this D-Pin is D-Pin is usually almost all D Pin projects you can think of, there's a trade-off that's being made. And the quality is lower, but the reach and deployability is higher. So with Helium, you can set up and deploy a hotspot for wireless and cell connectivity anywhere in the world for a few hundred bucks compared to the, you know, sometimes hundreds of thousands of dollars that it costs AT ⁇ T or Verizon to deploy something similar. The downside is that the reach is smaller. The range, it's a lower quality connection, but in aggregate, when you have millions of them, the performance is quite high. That's true of Filecoin, you know, IPFS, Rweave, any of these deep end systems, they tend to trade off quality for access, which is a quite reasonable trade-off to make. It's not a critique of them in any way. The difference with double zero is we are more performant than the public internet by far, but we have lower access and reach. The public internet is incredible. You can be, you know, on top of a mountain in Vermont and you can get public internet access. You're not going to be able to do that with double zero. And so D-Pin is definitely like from a technology standpoint, it's the right way to think about what we're doing. But our dynamic of a network is pretty different than a typical D-Pin network. Um, so in addition to having these fiber connections that are faster than what you can get on the public internet, there's a bunch of other technology that we're able to have on double zero that isn't available on the public internet either. And so two of those things are you know, signature verification and packet deduplification, um, which is basically the process of making sure, like when a when a signature verification is basically verifying public and private key cryptography. Does the signature on this transaction actually validate correctly when it comes in? And a blockchain is not able to include an invalid signature in a block. If it does, it'll be rejected by the rest of the network. And so we can simply go through and look at all the inbound transactions and use FPGAs, which are hardware that's reprogrammable on the fly, to basically verify that all these transactions are indeed good. And we throw out the garbage ones. Um, at the same time with deduplification of packets, you know, sometimes you'll get spam attacks on a network or someone sends a thousand copies of the same transaction. We can just throw out the ones that are identical there. And so these are kind of like DDoS prevention technologies that help the network be able to basically be more resilient and keep moving. But there's a whole bunch of other technologies we can use on this network that you can't use on the public internet. And one of the most important ones is actually called multicast. And everyone has probably run into the residential version of multicast in their daily life. It is how Sono streams music to multiple speakers. It is how AirPlay connects to multiple speakers, anything where you're doing a one-to-many connection on your home network. Um, Chromecast runs on multicast as well. All of these things basically use the network for hardware acceleration of packet replication. So your phone is not sending out 15 copies of music to 15 speakers in your house. That would overheat your phone in a pretty quick fashion. It's sending out one copy, and then the network is replicating that data to get where it needs to go. High frequency trading firms in the financial market have used this technology for decades. This is how they move prices around. It's much, much faster and much less resource intensive than sending out data one at a time. And so one of the really cool things with multicast is we can replace things like turbine and raptorcasts, and we can basically use this for state propagation. So to kind of take a step back, what happens in a blockchain? So my validator builds a block. That's great. I've got this block. I need to now get a copy of this data out to 2,000 nodes all over the world. How can I do this? Well, you have to send it out one at a time to each of those nodes, or you have to create this tree system where I send it to you and then you send it to someone else, and it's kind of a giant game of telephone. And all of that takes a lot of time. But with multicast on double zero, you can send that data once. And then the network automatically replicates it to everywhere it needs to go. And what's really important about this is the replication always happens as close to the edge as possible. So in an example where you have like a thousand validators in London and one in New York, on a traditional network running over the internet, that one validator in New York has to send a thousand copies all the way across the Atlantic to get to those validators in London. With multicast, it only sends one copy across the Atlantic. And then the switch locally replicates the data out to everywhere it needs to go. So it's a much more efficient way to use bandwidth, and it allows these networks to run at much higher data rates much more quickly than they would be able to otherwise.

SPEAKER_00 17:02

Very cool. I appreciate that. So one last question. If I remember right, you have something called a proof of utility. How is that tied into all this?

SPEAKER_01 17:11

Yeah, so we kind of had to come up with a new reward distribution mechanism for the double zero network because it's a very different type of network. So if you think about like a proof of stake network like Solana or Ethereum or pretty much every network out there now except like Doge, um they run on proof-of-stake systems. And uh the value of proof of stake is it's really good at securing value over the long term without using a ton of energy. And so the way proof of stake works is it's it basically says like if I have a you know $3 billion of USDC minted on chain, that needs to be secure forever. So how can I know that that's secure? Well, this validator building a block, it's got you know five billion dollars staked to it by itself and a bunch of other stakers and individuals. And so if that validator misbehaves, it basically has more to lose than it can gain from stealing. That's the whole theory behind proof of stake. The proof of stake is literally assets at stake and assets at risk. So I can trust you because I can slash your validator if something if you do something bad. Double zero doesn't actually have, it's a stateless network. No one's deploying funds on double zero. They're deploying Solana validators or SUI validators or Aptos validators or some other type of system. It this is an OSI layer one, two, and three network. It's really far down in the stack. Um and so because of that, we don't need to we don't need persistent security in the same way. So proof of utility actually much more closely uh mirrors proof of work. So the nice thing about proof of the downside of a proof of stake system is if you are a large validator and you have 2% of the stake of a network on it, you're gonna get a huge amount of rewards. But your validator is not doing like 4,000 times the amount of work that my validator is doing when I just have a tiny little amount of assets on it. So you're getting disproportionately rewarded for securing the network, even though you're not actually doing more work. You have more assets at risk, but you don't actually have any more like opex or capex cost associated with that. In Bitcoin and proof of work, if I generate 10% of the hash power, I'm gonna get 10% of the rewards. It's an incredibly linear relationship between security and reward. And so we wanted to replicate that in our model too. And so this is where this concept of proof of utility comes in. And so we apply these things called Shapley values, which is an economic concept where you basically say, how much worse would a system be if I take this piece out? It's like a Jenga tower, right? Like you pull out an individual block, you're probably fine. That individual block's not that valuable, but there's certain blocks that if you pull it out, the whole thing falls down. And so there's in in a in one cohesive system, different parts of that system have different values. And you look at those values relative to if they were removed from the system. And so we use Draply values to assign a value to every contributor in the system, and that's how we generate a reward rate. And so, you know, this what this has a few really nice properties. One of these nice properties is it encourages competition between um contributors. So if we have Galaxy providing a route and jump providing a very similar route, and the Galaxy route's a little bit faster, they're gonna get more rewards. And so that's an incentive for the jump folks to be like, oh man, how can we make our network route just a little bit faster to win back some of those rewards from our, from our collaborator, but also our competition. And so it goes back to the fundamental quality that made Bitcoin so powerful of weaponizing greed and capitalism to produce positive sum outcomes. I want to find you misbehaving because I get to steal your block rewards if I'm if I find that. And so it prevents collusion and it's this really beautiful economic system. And so proof of utility is our attempt to replicate that on a stateless network that doesn't have to secure assets over a long duration of time. We need is momentary security and we need as much utility out of each network link as possible.

SPEAKER_00 21:23

Very cool. I love hearing it. I don't know if you could see it. Like I'm getting this smile just hearing about it, because the elegance of like a beautiful of an economic system that aligns everyone is what like kind of gets me got me into crypto in the first place, right? It and like you, I'm glad you mentioned Bitcoin because it it's about uh replacing incumbent systems with new ways of doing things where everyone's more aligned. And and I just love it.

SPEAKER_01 21:47

Um Yeah, we have all of these like I would call them like almost like mind viruses in crypto nowadays, where we just we keep putting band aids on systems and saying like, oh, a good actor will do this. This is like there's a lot of incredible things in the Ethereum ecosystem, but this is incredibly prevalent in the Ethereum ecosystem of like, what if we just created MEV boost and producer-builder separation? And uh, oh, these things aren't able to actually make money. Well, we'll just subsidize them from the Ethereum Foundation. And it's like, okay, guys, I get it, but like systems that require subsidies are fundamentally unsustainable. You have to have these systems built in such a way that people can have a profit motive to improve them. And it's a very difficult system to build, but when you can build it and it works perfectly or it works good enough, you start to create these really powerful economic flywheels. And I worry that we've sort of lost sight of like greed and capitalism are very good things in crypto, as long as we can weaponize them towards positive sum outcomes. And uh that for us is just like that's the ultimate like North Star from an economic design standpoint is can we produce better outcomes while making people more money? That's it. And if that money comes at the expense of someone else's economic output who has designed a worse system, like tough luck. Like it's almost it's a very sort of like uh laissez faire, like my factory is marginally better than your factory, so I steal all your customers kind of approach to it.

SPEAKER_00 23:23

I can't man, I'm all I'm all about it. Um so you we discussed a bit about validators. I just you know it's part of the still going over uh some of the basics of uh double zero. Could you explain to us how does joining the network work as a validator or as a fiber contributor? Um and what you know, what changes for them like once they join? And then what happens also like if the their connection to the double zero network drops? Yeah. Can you just help us explain intro to validator and double zero one on one?

SPEAKER_01 23:52

Yeah, totally. So um we run into a bunch of terminology problems because crypto, as it always does, reused a bunch of words that mean different things in different industries in its own way. So one of the examples here is we have a network, and our network is not the Solana network as sort of an abstract concept, it is the physical fiber connecting all of these switches. And like when I say physical fiber, like I think people don't always realize like I mean these are actual cables on the floor of the ocean or running over mountain ranges, and everything is actually plugged into everything else. Like it really is like a when you're a kid and you've got like two cups on a s connected by a string and you're playing telephone, like it is a giant network of interconnected systems. Um, and so validators for us usually refers to like the Solana node that's connected into the double zero network. Because our fundamental unit of contribution, actually, let me take a step back. The only reason you have a validator on Solana is because something needs to provide compute. Like at its at its core, all the validator is doing is running a piece of code that creates a virtual machine, which is basically just a giant software abstraction version of a computer that you can run programs on. If you didn't actually need to run the programs on it, you wouldn't need validators. This is why Bitcoin has miners as opposed to validators. They're not a Bitcoin miner isn't really processing data. Like the the actual processing of user transactions on Bitcoin is like a billionth of a percent of the compute usage of a Bitcoin node. The majority, basically all of a miner's usage is cracking this algorithm to be able to earn the right to build the next block. That's different on Solana. 90% of a CPU's usage on Solana is running the virtual machine and actually executing the program and filtering through transactions. Its fundamental unit of contribution to the network is compute. On double zero, our fundamental unit of contribution is bandwidth. And so we don't need validators in the traditional sense. What we need are 100 gigabit fiber networks hooked up to high performance switches and FPGAs and all this very specialized appliance computing that's in line. And so most of so validators operating on double zero, they're not like building the double zero blockchain or something like that. We do have that, and I'll get I'll get to that in a minute, but they're mostly using it as, you can think of it as using it as their ISP. So typically, if you have a validator sitting in a data center, it's got an Ethernet port or four Ethernet ports on the back, and that just sort of talks out to the public internet. And so we instead have is we have some of those ports talking into the double zero network. And so the validator can now broadcast to, I want to send data over the public internet, and there might be stuff that's not reachable in double zero network. If you've got to go grab a software update, that's gonna be from the public internet. If you're talking to a user because you're an RPC server and you're talking to a phantom wallet, well, my phantom wallet's not operating on the double zero network, it's operating on my local computer. And so it that server still has to receive data from the public internet. But then everything it can talk to on the double zero connection is like a higher quality, faster network. So we do also have this underlying thing called the double zero ledger. And this is technically a blockchain, but it's not a blockchain that a user would ever deploy anything to. And we kind of get asked sometimes, like, oh, why are you using a blockchain? It's like, well, it's the most perfect system of accounting humanity has ever come up with. A blockchain is basically an unfakeable, unhackable database. Um, and so we use that to do things like post-telemetry, um, have all of the user accounts asserted on it. And so we do have this role of what's called sort of a resource provider. And those are folks who are running a basically copy of the double zero ledger, and that's doing a bunch of things to like prevent contributors from cheating. They're sort of enforcement agents that sit on the network and they say, hey, like these fiber links have a latency of you know 125 milliseconds. This data took 130 milliseconds to get through. We should go investigate if someone is trying to interfere with traffic or something along those lines. And so they basically run sort of the enforcement component of the network. And that's really important for making sure that the network is censorship resistant and trustworthy. You asked a question too about like what happens if a piece of fiber goes down. Well, this actually happens all the time on the public internet. Fiber gets cut all the time, either it goes down for scheduled maintenance, or maybe it gets dragged by a ship anchor. Sometimes there's some Russians involved. There's all sorts of reasons that fiber uh goes offline. Um and so we have the same problem on double zero. And the only solution to fiber going offline is multiple redundant pathways. And so we have this concept that we're building out with our contributors of three rings around the world. And you can think about like one sort of northern hemisphere ring, one sort of equatorial ring, and one southern hemisphere ring. And so not only do we have, you know, six pathways to get from New York to London over that northern hemisphere ring, but if that's offline, you can go down, you instead of going from London, your data can go down through Marseille and then can connect in to like Ashburn and Virginia as opposed to coming in through New York. Um, and that's sort of the this this real resiliency and redundancy over the long term comes from the ability to have multiple pathways in which data can flow. And on the public internet, most of these cables are run at 90 to 98% capacity. They're trying to jam as much data possible into the pipe because every time you're not using the pipe, you're losing marginal cost. The incentive on the public internet is to, it's just like a, you know, a FedEx truck or something like that. You want to pack it as tightly as possible. Um the problem with that is it means when there's an issue, which happens constantly on the internet, there's no excess capacity. And so you'll see, like we had a cable, we we sorry, we didn't have a cable. The world had a cable that was down between um the west coast of the United States and Tokyo for about a month and a half. And that cable was down for scheduled routine maintenance. It had been posted about for you know six months ahead of time. They're upgrading some systems on it. If you ran a validator in Tokyo, why would you be monitoring the alerts for sub-C cables that are going offline? And suddenly there was this moment that happened a few months ago of a bunch of validators in Tokyo were like, oh, we're having trouble keeping up with the network. What's going on? Turned out the data center they were in was using this cable. And that data center did not have peering agreements with other cables that went off that way. So the data coming from Tokyo, trying to get to like the United States, was actually going the long way around the world. It was going basically down to Singapore, over to the Suez, and then over across the Atlantic. And, you know, it was, it's all these types of things where you don't think about this when you're a when you're just a Solana validator because you shouldn't have to think about it. The thing is, blockchain's fast enough now that you do have to think about it, or you have to use the double zero network. And that's kind of the real thing we're trying to do is we're trying to bring this much higher level of performance to blockchain without making validators suddenly have to worry about like ship anchors and cable maintenance and all these other types of things.

SPEAKER_00 31:35

That's really cool. That's really cool. Um, so I feel like we've we've we've really thoroughly given everyone kind of a level set on what double zero is. I'm really curious. You know, you've been in testnet for a bit, mainnet's coming up soon. Um I don't know if if exactly when, if you could tell anyone, but um, you know, if uh I'm curious, when you go from testnet to mainnet, who's using it on day one? Like what's the demand for this?

SPEAKER_01 32:00

Yeah.

SPEAKER_00 32:00

Yeah.

SPEAKER_01 32:01

So at the moment, um, so our testnet has been up since we launched it in late February. And testnet only consists of eight fiber links, and there are 10 gigabit per second fiberlinks, and that kind of makes a loop around the world. It's also about only one provider. We did not expect that to become a place that people would run mainnet Solana validators, but today about 16% of the Solana network is actually running on the double zero test net uh as of September 15th, where we're recording. So uh we did not expect that to happen. So we were accelerating the launch of mainnet to basically get people off of testnet so we can go back to using it as something we can break. Um, and so our our mainnet launch is coming up in the next few weeks, uh, which we're very excited about. And the difference here is we go from eight fiber connections to 45 plus different fiber connections. So we're massively, and we'll be at 75 by the end of the year. Uh, we're massively expanding the number of data centers we have access to, the resiliency and redundancy of these systems, and also many of our new fiber lines coming online are 10 times faster than the current ones in terms of capacity, and they're even faster in terms of latency reduction. So we haven't actually talked about this component, but like there's kind of three pieces of double zero that make it different than the public internet. The first we talked about, which is we can use technology that just isn't ex that does not exist on the public internet. And it doesn't exist because of some like technological barrier. It's business model problems. You have hundreds of ISPs in the world and they all fight with each other about standards and who bills what and who gets paid for what. So no one can agree that they want to use this like new technology. That's a that I say new technology. It's existed for 20 years and the public ISPs still don't adopt it. Um, the other one is faster pipes, which we or fatter pipes, which we talked about already, which is the idea of like most validators on Solana today have somewhere between a one gigabit and a five gigabit connection. Double zero's connections are usually a hundred gigabits. So we're talking about anywhere between 20 to 100 times more data throughput capacity. And the next one is latency and jitter. And so latency is if you grew up playing games in the 90s, you in you intimately know this value. It's your ping time. It is your latency between your computer talking to the server. And if you have bad ping time, you're gonna miss your shot half the time when you take the shot. If you have good ping time, you're gonna actually be able to front run, to use financial terms, the other player, and you're gonna be able to get the headshot off or you know, whatever you're doing. So with the public internet, you have a lot of latency caused by indirect routing, but you also have this quality called jitter. And jitter is a measurement of how much the variance in latency changes. It's sort of like acceleration versus jerk in physics. Um, and really what it's about is like if it always takes me 200 milliseconds for my server to talk to yours, that's a long time, but I can model that, I can do a lot of things around that. But that's not actually how it works. What happens instead is sometimes it's 120 milliseconds, sometimes it's 400 milliseconds, sometimes it's 300 milliseconds. There's this huge amount of variance that happens when you're using the public internet. And that's fine if you're loading websites, that's fine if you're watching Netflix, right? Netflix is all buffered data, so they can handle that sort of jerkiness really easily by just using more of your RAM to store the TV show that you're watching and preloading it. Doesn't work on Solana, doesn't work on high-performance systems. And so not only will double zero reduce the latency, meaning we actually have pathways that the fiber runs on that are much more direct and much faster than the public internet. So we're talking about, you know, over 100 milliseconds of latency reduction across some of these routes, sometimes 200 and 300 milliseconds of latency reduction. But if it's, you know, if the system says 137.5 milliseconds to talk from this server to this server, it's gonna be that same time 99.9% of the time. And any jitter you get is incredibly small compared to what you get on the public internet. And so you can think about this from a trader perspective. If my systems tell me, hey, we th we have a high degree of confidence in the next 150 milliseconds, this is the price the market will be in. Well, if I know it takes me 125 milliseconds to talk to the exchange, I'm gonna make that trade because I know my execution time, I know my estimation time, and assuming my quants have done their job well, I know I'm gonna be in the money on this trade. If that transit time could be 120 milliseconds or 170 or 200 milliseconds, suddenly you have to decide, well, is my economic opportunity worth making this trade or not? Because it's possible that I send this trade and by the time it gets there, the market has already moved past what my model can predict. And so would that be good for me? Would that be bad for me? Like what is my risk profile? And suddenly it's a much more complicated calculation. And so our whole goal with double zero is to bring much more predictability to networking that that'll benefit protocol developers, and that'll also really benefit traders and market makers who are operating on the networks as well. That's awesome.

SPEAKER_00 37:25

Um, on that note, then are there any early metrics you could share, kind of like pre-mainnet? Or are you still running experiments?

SPEAKER_01 37:32

We've learned a bunch from testnet. I think what we've seen is like, you know, uh timely vote credits, which is a SLANA-specific metric, um, is down. We generally see there's this thing called skip rate, which is how often a validator fails to build a block when it's supposed to, and that's usually due to data not getting to it in time. Um, we've seen validators have, you know, pretty intense reductions in their skip rate when they're operating on double zero as well. But, you know, double zero in its test net form at the moment, you know, it's only eight fiber connections. And so there's limited places that were to actually like strongly applicable. What we see are like ping times are down a lot. But in terms of you know, actual economics, we have some validators that are reporting they're making more money than they were before, and others just simply saying, hey, our validator is like more caught up than it's ever been before. So we don't have hard really hard metrics to share in part because testnet is a limited testing experience. Um, but as we roll out mainnet, like we'll have all tons of dashboards available for people to be able to see in real time just how much performance improvement they're getting over the public internet.

SPEAKER_00 38:39

Very cool. Um well, since we've kind of gone over uh some of the supply side stuff and like kind of the evidence that that this stuff works, um, would love to maybe dive into the economics and the incentives, right? The token, uh the contributors, et cetera, just a little more in depth. Could you please break down the tokenomics for us? Like how does the double zero token focus on the ecosystem? Um and yeah, let's let's start there. Also, what will the name of the token be or the ticker symbol or anything if you've decided on that?

SPEAKER_01 39:09

Yeah, so the the the token is a 2Z token, so the number two followed by the letter Z. Uh and the way this token functions is as a payment token on the network. And so if you're a validator and you're operating on double zero, you pay a percentage of your earnings to basically get that network connection in. And so, you know, from that perspective, you might be paying, you know, 5% of block rewards on Solana or something like that to be a part of this of this network. And our our promise to you is that we're gonna, you know, our North Star here is to make validators more money. And if we're not making validators more money by helping them build more efficient blocks or better MEV opportunities, um, we're not really doing our job or or allowing Anza and Fire Dancer to massively increase block space, which also increases revenue for validators. And so that's kind of our north star is like, can we make validators more money in aggregate? Um, and so they pay a fee into that. That is then swapped into the native double zero token. About 50% of that is burned. You can check this all out on our economics blog. Um, and the other is distributed out to the fiber providers according to those Shapley value proof of utility numbers that we we talked about. So these will change every epoch. So for example, if you it might be that like one epoch, your switch has let's say two and a half percent of Solana stake located that it's serving. But then a new data center comes online, connect to double zero, and you go, hey, like I actually want to move because this data center has better pricing, or you know, I would like to run in my in, you know, my own country. And double zero has now brought connectivity down into Buenos Aires, and so I can actually run a validator in country, and I love my country and I want to run it here. And so that amount of stake in various places will change epoch by epoch. We see people moving data centers for 10 basis points of additional MEV. Um, and so every epoch your reward rate will change based on the utilization of your network links and how much value they're providing into the system. So it's a very dynamic system.

SPEAKER_00 41:16

And I'm curious for uh people who aren't necessarily in crypto, like you mentioned, gaming, CDN networks, AI workloads, TradFi HFT. Um, how will how will they pay? You know, because it is that like a percentage of yeah, how does that work?

SPEAKER_01 41:33

So there will be kind of a more of a like a pay per request model similar to a UCN RPC providers for folks that don't have on-chain revenue. Um, you know, a lot of that stuff is is bespoke because the data workload, like if you're looking at like a protocol like Solana, it's very easy to say like the amount of bandwidth a validator needs scales proportional to their stake weight. Like they're that's very easy to look at that and calculate that and figure out what that is. It's it's different with RPCs, right? Some RPCs are time optimized and they're really fast response systems and they charge people a ton of money, but they don't send that many requests. They're mostly meant to be like, you know, quick response force type things for like arbitrages. There's other validators systems or there's other RPC systems that are meant to serve millions and millions of generic requests. And so we kind of have to change the model a little bit as we deal with more bespoke systems. So our first non-Solana use case is something called Shelby Storage, um, which is a storage project that Apdos and Jump have teamed up to build. And it's it's hot storage. So it's not competing with IPFS or RWEV or something like that. It's competing with AWS S3 and Cloudflare C2. It's meant for data that is read often. Um, so data that's read often can be like live sports, TikTok type products, you know, SQL databases that are hit very often. It's not kind of your like archival storage of like, oh, I need to dump, you know, like a hundred years of like a newspaper's archive onto a system. IPFS and such are great for things like that. Um, and so for that, what they really care about is fast delivery of data and fast response times. And that's where kind of double zero comes in. Those data models are very different than Solana. And so we have to kind of build an economic model that's slightly tail uh tailored to each need. And in that way, double zero is a little different than other types of programs as well. Because, you know, on Solana, it's not like, oh, there's a special, there's there's a well, there is a differentiative fee for votes. But apart from votes, um, there isn't sort of like, oh, an NFT project pays a different um rate than like a DeFi project or something like that. Um, but in these cases, there's different things that different networks care about. And so we do, we also have different classes of service. And so we can. Customize this to meet the needs of a protocol to sort of make sure that those protocols are both appropriately contributing to the protocol revenue of double zero, but also that we can service their needs in different ways. And so we're kind of doing something new. I'm not really aware of other protocols that are both that are not centrally run, but do have this kind of uh variance and adaptability based on how they're used. Like Akash is a great example, right? It's a compute network. It kind of builds you the same no matter whether you're running a Solana node or an app or an Aptos node or an Avalanche node, um, even though those characteristics are different. So I think there's a lot of stuff that we're kind of trying to push the bounds on for how protocols think about um, for lack of a better term, customer segmentation.

SPEAKER_00 44:50

Yeah. No, I think it, I mean it makes a lot of sense, right? Like if people need sports data fast, that's very different from needing pricing data on something or if they need or something else, if they need something frequently, yeah. It's really cool. It's also really cool to see that you guys are already working on solutions that are going to affect uh more than just you know the Solana blockchain, that you've got more than one kind of like initial target audience. Like, I think that's that both that's really cool. Um thank you. So I'm curious, we were talking about bandwidth. So to access bandwidth, is there like a stake to access model or like a whole 2Z model?

SPEAKER_01 45:26

Yeah, like what's some of the utility for the payments or so you have to think about this from the perspective of what we're trying to accomplish. And a lot of the philosophy for this comes out of the four years I spent working with Anatoly at Solana. There is a scarcity mindset to digital resources that pervades blockchain. And the only place it makes sense is Bitcoin. Right? Bitcoin is meant to be there's only 21 million of them for an arbitrary reason, but that creates certain economic structures and systems. You don't want Solana to have scarce block space, you want to have extremely abundant block space because it's you can kind of compare this to like you're either like Lockheed Martin or you are like DJI. And so Lockheed Martin is doing like massive cost contracts to very few buyers for very small units. It's like, oh, they're gonna build like a thousand fighter jets and each one's gonna cost $400 million, or you can be like DJI and you should be like, we're gonna sell $400 drones to literally every person on the planet. And those are equally valid business model approaches. But when we're dealing with software systems, the limiting factor for those software systems is just what artificial caps you put on them. And so the problem with a staking model, in my view, is that it it limits your TAM pretty quickly. You either have to keep adjusting down the amount of stake you need to participate in the network, in which case, why bother having the staking component at all? But it also sets up pretty meaningful barriers for traditional systems to engage. Um, pretty famously, Visa and other companies that operate crypto payment rails under the prior administration thought it was too risky to hold gas tokens on their balance sheet. So we had this whole convoluted system called gasless relayers that came up, which was like, oh, I can pay transact, I can pay my, instead of paying, you know, 10 cents in Seoul, I'm gonna pay 12 cents in USDC, and that's gonna swap to Seoul on the back end to pay for the gas transaction for this thing, because these companies felt it was too regulatorily risky to actually hold, they thought it was too regulatory risky to hold Sol or Ether on their balance sheets. Plus, do you have to deal with suddenly like the tax accounting components of this of like, oh, we pre-bought, you know, 10,000 sole at $20. It's now $200. Do we have a giant capital gains tax problem every time we pay gas transactions? There's a lot of these problems that are non, they're they're not problems at all on the user level, and they're not even problems on the crypto company level, but they become pretty massive problems when you're dealing with a big corporation, especially a publicly traded one that's got to do gap accounting. Um, and so we thought it was really important that folks are able to show up with nothing except a validator and pay for transaction fees on the network. And we we will have systems in the future, or I'd I'd love someone to build them out, otherwise we'll build them out ourselves, that basically are the same type of idea of a gasless relayer. But because you don't have to pay for transactions uh in the same way on double zero, we can have prepay balances, which is just like, hey, I funded a bunch of balance in this account. It's paid down over time as I use the network. Um and so we have a lot of these models that I think are are are gonna be much more friendly to non-crypto users as well.

SPEAKER_00 49:01

Super cool. I like that you guys are thinking of solving that problem for people and that and who else would know about what these like big companies are thinking, other than you guys. Like it's it's I love that your experience before gives you the guidance and insight to avoid pitfalls today. Like that's that's really cool. Um, well, I'd love to jump into you know a little bit about what what makes crypto really cool, which is like fairness and decentralization and governance. Um I'm curious uh for Solana validators who aren't using double zero, how might their performance and economics be affected as the network grows? Because I'm curious, like, will there be like this two-tier network?

SPEAKER_01 49:38

Um yeah, I mean that's entirely up to Solana. So we kind of had this like this moment in one of the like the core dev meetings where there was a bit of a question about like, do we want Solana to go faster than the public internet? Which like you you kind of think about it one on one hand, you're like, yes, of course, as fast as possible, like go, go, go, IBRL. On another hand, you're kind of like, well, maybe. Like, do do we want to say that you know, if someone wants to run like that? This has sort of been the critique of Solana from the Ethereum folks since day one, is like you it's hard to run at your house. Fair, right? You you need a good amount of bandwidth to run a Solana validator at your house. If we adopt double zero, uh okay, suddenly you've gone from needing one gigabit to maybe needing 10 gigabits at your house. You can basically only do that in South Korea and like a few a few other places in the world that have incredibly fast internet. Um so uh, you know, is this something that the protocol actually wants? Do we want to say that like if you want to be a validator, you need to exceed the speed of what the public internet is capable of? And the answer really came back as yes. But that is like a real philosophical moment for a network to say, um, do we want to exceed the public internet speed limit? Or do we want to be stuck with the public internet speed limit? Now, Ethereum chose like, no, we want someone with a one-megabit connection and a shoebox to be able to run a validator. And we don't want Ethereum to go any faster than this. I think that was the wrong choice, uh, personally, but it was a choice that was made. And so this was this is the kind of a whole question I think every protocol that adopts double zero is going to have to ask themselves is do we want to push the bounds of what's possible, or do we want to say access is actually much more important for us than performance? And uh I don't have a prescriptive version of what protocols should make. It's up to each one of them to make their own decisions. What I would say though is I think the Bitcoiners are fundamentally right. The Bitcoiners are right in that true censorship resistance from governments, from corporations, from anyone, you have to be able to validate blocks on an FM radio broadcast from a satellite using having no internet connection. Or you have to accept that we live in a society. Like you can sort of think about it as Bitcoin is almost impossible to shut down. And it's almost impossible to shut down because you only need a few kilobits per second of data to actually validate Bitcoin blocks. And you can get that data over a 56k modem, a ham radio, an FM radio broadcast from space on a satellite. There are block block data for Bitcoin is actually broadcast from a satellite in space. So you can actually validate completely offline in like the wilderness, you know, in your off-grid cabin with your Bitcoin miner if you technically want to. If the OECD countries really gang together and want to shut down Ethereum, they're gonna be able to do that. And, you know, people sort of talk about like it's really important I can run a node in my basement. It's like, okay, but how many ISPs do you have? Because I guarantee you Verizon has the technology to do a deep packet inspection and figure out that you're running an Ethereum node and block those connections if the government orders them to. China has proven that the great firewall is pretty great and it really can be a quite powerful um tool for blocking certain types of data. Now, we live in a you know Western democracy, and it that type of thing doesn't happen here, but that's not the way you should be thinking about these technologies if you truly are thinking about like, you know, the rise of authoritarian regimes and World War III and true censorship resistance. Like the Bitcoiners are totally right about this. So, my kind of view has been anything like the minute you pass the Bitcoin level of data, just accept it and go full speed ahead. And there isn't a lot, like everything else is kind of mid-curving it, in my view. And it doesn't mean you have to use double zero, but you should be very comfortable exceeding what's possible on the public internet, or at least exceeding what the majority of household connections can have, because otherwise, we're never going to be able to really compete with TradFi. And that is the goal at the end of the day, is to change the fundamental way that these systems run, to make them more fair, to put control back in the hands of people who actually run and manage and use these systems, not people who have been arbitrarily granted a license by a government to make money. Um, and that is sort of the thing we're all after is open and fair competition. Uh jump trading on double zero will have, you know, using double zero to trade on, you know, Solana and Binance and all these other places will have no advantage over the average user coming in to trade on these systems. That's not true in TradFi. Like you cannot, if you are the smartest quant in the world as like a 14-year-old, you are gonna you cannot compete on the traditional exchanges because you simply don't have the money to buy the license to trade on these exchanges. And so you're always gonna be running on slow data. That's not true on chain. And I think that's a really important thing to highlight, and that's the way we want to move these type of systems uh into the future.

SPEAKER_00 54:58

Very cool. Um, uh the just a follow-up question on that. You mentioned that jump and the average user would have similar like uh bandwidth propagation like capabilities here, uh data propagation capabilities. I'm curious, how does the average user though know that he or she is somehow using double zero? Like, will validators have a little badge next to them, or do you have to just go look at the list and somehow work your RPC to this validator? Like, how does that work?

SPEAKER_01 55:27

So on validators.app, there actually is a little badge. Um, just like you can see a Jito badge or a Fire Dancer badge, there's a double zero badge. You can actually go look up who's running on the network. But this is part of why we have this um, you know, double zero ledger as well. It's a publicly accessible, publicly queryable database. It's a blockchain. Uh, and so you can just query the explorer and you can say what validators are connected to double zero, how long have they been connected, what are their performance characteristics, and you can go and you can rent a box in TerraSwitch or Latitude or Cherry servers, and then you can just connect in to double zero. And you know, yes, you still have to go rent a server, but like that is significantly cheaper than a multi-million dollar uh, you know, speed feed from the stock exchange. You're talking 700 bucks a month, like that is real money, but if you're you know actually trading, it's not that much money uh compared to especially what you need to actually start trading truly on a stock exchange.

unknown 56:27

Yeah.

SPEAKER_00 56:27

Also, when you think about like the tools that you're giving people to then make more money for themselves, it's it's awesome. Like I really, I really love it. I love the fact that you're giving entrepreneurs, but also corporations and like businesses the opportunity to like do more and be more productive. And that's huge because like you mentioned, you know, Bitcoin was great in that it brought out censorship resistant money and a censorship-resistant ledger. Um, then we had Ethereum which gave us programmable money, and now we I want to call like we have actionable money, like you can like take action and and be productive with it, which is something you just can't do with you know on a Raspberry Pi. Um, so I'm I'm cool with that. Um I'd like to just talk a little bit more about uh uh governance stuff. So are there any like anti-concentration policies? Um, or is that or is that kind of solved by the proof of utility slash proof of work inspired systems?

SPEAKER_01 57:18

Yeah, so we take a pretty um neutral, so like we describe double zero as base layer neutral infrastructure. And so if there's nothing preventing co-location or concentration of, let's say, Solana nodes or SUI nodes or whatever on double zero, that is up to the protocol itself. Now, the difference is right now it's actually very hard to get geographic data on where a node is is run, um, because people fake their locations all the time, you know, because they want to appear like they're somewhere they're not, all these types of games. Because we know the latency of every connection into double zero and where these switches are located, um, you know, something like the Solana Foundation can just go on double zero and they can look up where all these validators actually physically are. Now, we may not, you know, it's not like, oh, it's in this rack in this data center, but if we know you have like a one millisecond of connectivity into this node, we know you're in that data center or you're within, you know, 500 meters of that data center. And so it's then up to the protocol itself, the actual blockchain to incentivize decentralization or penalize centralization. So you could see, you know, the Solana Foundation delegation program saying, hey, we're not going to delegate to anyone in data centers that have more than half a percent of stake in them, or in cities that have more than 1% of stake in them. You could you can you could see for the first time these protocols having tools to be able to incentivize people to stake and to run nodes elsewhere, but I would all consider that above the protocol, above the double zero protocol. And we don't take any prescriptive actions on how folks should. There are networks that are intentionally centralized, right? Hyperliquid is intentionally running in one data center in Tokyo. And so who are we to tell them, like, no, the protocol is going to penalize you for this any more than we should tell Solana, no, you should run everything in just a few data centers. Um so we really approach this from like a neutral perspective. But if you meet on the fiber contributor side, um the way that works is basically we we do reward redundant, resilient connections into the same location. But if there's eight connections into one city, the reward per connection is just gonna be lower. And so, you know, natural human instinct of profit-seeking motive will say, hey, if I can make more money putting a fiber cable in in a different location, I'm gonna do that instead.

SPEAKER_00 59:45

So I'm curious, what becomes like permissionless when? Are there any like decisions that you hope to decentralize over time that but today our foundation stewarded? Um and can you give us any insight into that like governance roadmap?

SPEAKER_01 59:58

Yeah, so today we have 11 independent network contributors. Um, and joining the network once we go to mainnet beta will basically be anyone who is in the Solana leader rotation will be able to join. So this is just a civil resistance mechanism as we get started. You have to launch your validator on the public internet first, and then you can migrate it onto double zero. Uh anything that's basically in gossip is the way to think about that. Um, over time, we'll be decentralizing more and more components of the protocol. This is why we're calling it mainnet beta. There are a few components that are centrally run at the moment by the Malbec Labs team. But again, this isn't a stateful network. And so, you know, we're not storing user funds on it, we're not storing tokens on it. The token is an SPL token on Solana. Um, and so what I would say is all of the kind of core economic components, um, there's no sort of theft of funds ability or something along those lines because of the way that we've we've built the system out. And so, like, we will get more people running the double zero ledger over time, uh, and then building up this sort of resource provider community that's doing these sort of checks and balances on network contributors, uh, we'll definitely be expanding that um system as well. So, you know, our road from mainnet beta to mainnet, I imagine it's somewhere around a year. Um we still have to figure out, you know, what exactly that looks like and how we expand out from there. Um, but you know, uh decentralization is very much a big part of why we're building and doing this thing. So what we wanted to start with was sort of the most important components of decentralization, which are multiple independent contributors and full transparency and auditability around how data moves and the telemetry of things happening on the network. And then we'll move from there. So, right now, for example, the system that decides if you can join double zero or not is a centralized system at the moment. Now, that doesn't actually matter much because again, not being able to join double zero uh if that system is offline or something like that doesn't actually impact the double zero network itself. These are all components that we have to uh decentralize over time.

SPEAKER_00 1:02:09

Very cool. Um, so I'd love to talk a bit about partnerships, growth, one, three, five-year goals. Um, can we maybe start with you know, who are your key institutional or business and community partnerships so far? Um, you've mentioned that some of your investors have been contributing to fiber. Are other people contributing fiber? Has a community reached out? And like, are there any community efforts around that? Yeah, I would love to talk about partnerships.

SPEAKER_01 1:02:34

Yeah, so contributing fiber is the most valuable thing we have at the moment from like a core network support perspective. Um, there's a mix of folks who are doing that. Um some are investors, some are not. Some are simply validator operators that want to make their data center more attractive. And so they're bringing double zero into their data center as a way to sort of attract new customers to that data center as well. So there's a whole myriad of reasons that people are contributing fiber to the double zero network. And that's one of our core components here is like we want a whole different type of use case involved in the network. I think that's where true decentralization doesn't come from the number of nodes you have, the number of validator clients. It comes from the number of sort of different resources involved. The classic version of this is like, you know, USDC will counterbalance um other incentives on like the Ethereum network, if there's ever a question of a fork or a rollback, or if USDC gets hacked, I don't think the Uniswap community is going to let Ethereum roll back the chain and undo the hack, right? That's where the true like Rock Sala decentralization of these things comes from. Same, same on Solana. Um so you know, as we as we move out and onboard users, users being validators, those are obviously an incredibly important component of the network as well. Um and so, you know, over time we will start to support non-blockchain uses as well. But our our our first goal is to nail the Solana use case because Solana's the hardest. And so uh, as you know, Anatoly always says, do the hardest thing first because then everything else is easier.

SPEAKER_00 1:04:13

Facts. Facts. Okay, cool. Um uh then on a let's say multi-year basis, are there other uh categories of contributors or people or institutions or community members that you want involved? Like do you want an actual team code involved? Do you want actual like stock exchange involved? Or yeah, how would or CDN?

SPEAKER_01 1:04:35

Yeah, I mean you kind of you kind of fast forward and we would love connections into all the major both crypto and stock exchanges. So people can actually send like API trade requests to Binance and OKX and Bybit and Upbid and Coinbase through the double zero network as well. Um, we think that's really big for you know arbitrages and other types of trading activities, which kind of goes into one other piece we haven't talked about, which is next year we will be launching something called prioritized data. And so, as I mentioned, there might be five different pathways to get from Tokyo to Singapore or you know, Tokyo to Chicago. They all have slightly different latencies and they'll all be completely fast enough for any blockchain consensus to run on. But if you are a trader and suddenly you see like a hundred thousand dollar arbitrage opportunity, you will pay more money to go over the fastest link possible. And so that's another big uh component of the protocol uh, you know, that'll be launching probably middle of next year. We'll see exactly when we have the whole thing built out. Um but that's another really big component of that as well. Um looking you know, longer term, we would love to replace IP addresses. What do you mean?

SPEAKER_00 1:05:45

With well, like you still need something that resolves like to a specific address, don't you?

SPEAKER_01 1:05:50

Like, what do you mean by you do, but if I'm trying to talk to like Delphi's validator, I don't really care what IP address it's on. I care about the cryptographic identity. Of the thing I'm trying to talk to. And so, you know, over the long term, we intend to work on public-private key routing protocols that will instead of routing you to an IP address, they'll route you to a pub key. Because we think that is probably the most durable and uh secure routing system someone could build. Uh, especially this allows us to do things like replace DNS, which is an incredibly centralized service. We can replace certificate signing authorities, which is another incredibly centralized service at the moment as well. Um, most major hacks that happen are that someone gets access to an SSL certificate. Like this is Stuxnet, right? Stuxnet was uh someone breaking in and stealing the signing authority from real tech to be able to install drivers that that tricked the Microsoft operating system into saying, oh, this is a trusted third party. It's blessed. Well, same thing can happen in networking, and you shouldn't be able to do that. Um and so you know, if we can switch things over to public-private key routing cryptography, um suddenly we get to a place where you cut out all intermediary trust certificates and trust authorities necessary in securing these types of systems. Um and so we think that's a pretty awesome thing for the long term to be working on.

SPEAKER_00 1:07:17

That's super cool. Well, I see, I love I love this ambitious vision. We're we're going for it. Um we're going for it. Um, I'm curious, you know, with with a big vision, uh, with with disruption, sometimes there's some certain types of risks. What's like a big risk or a couple of the biggest risks that keep you up at night?

SPEAKER_01 1:07:36

You know, I think a lot of it is around adoption and momentum. Um, we have really strong adoption so far, really good momentum. But if we get stuck at like 62% of the Solana network running on double zero, it's gonna be hard for protocol developers to say, well, we can now jack up the limits of the protocol. If we're up at 85%, it's kind of a no-brainer where they can just say, okay, we'll we'll start pushing the bounds of faster than the public internet can handle. So it's it's the classic, like, I wouldn't call it a cold start problem, but like the cool start problem that any new technology has in the space. And part of this is because we have a mix of validators. We have some that are very long-term minded and they're willing to join double zero. And we have others that are like kind of zero-sum mindset. And, you know, they're sort of like, oh, well, if you charge a fee, that means I'm getting less money. Or, you know, even if your fee is lower than uh like if if I'm making them 7% more money but charging them a 5% fee, they're sort of like, oh, I think I can probably do something like that on my own. And so it really depends on like, can we bring community together for the long-term success and adoption and advancement of the network in a way that will make everyone more money? It's a little bit of a governing the commons problem at some point. Um, so I don't think we have to worry about that, but I'm certainly conscious of that because I do think it's an important component to think of and to sort of keep in mind when we're dealing with any of these types of systems.

SPEAKER_00 1:09:08

Okay, cool. I appreciate it. Um, you know, I just want to do a quick couple, couple lightning round questions to kind of spice it up a bit. Uh before we wind down, I'm curious, like, what do you think is the most misunderstood uh what is the most understood thing about double zero? Or sorry, what is the most misunderstood thing about double zero?

unknown 1:09:27

Yeah.

SPEAKER_01 1:09:27

I think people haven't internalized yet that it really is a protocolized system, but we're not running the fiber. Um, I think because it's so unlike anything that you've ever seen before. Sometimes people are sort of like, oh, well, can't you just do this? And they're like, no, it's not our fiber. Like, it's up to the contributors. If the contributors all agree to this thing, totally, but we don't have any ability to force something through uh any more than Solana has an ability to force something through. Um, so that that's still an area we get of people not quite understanding how it works. This most often comes when we're dealing, um, you know, uh there's all sorts of places that this comes in. But um the other one that I'd sort of say is like there's a because people don't know how networking works, sometimes they think that what we're doing is somehow riskier than the public internet today. Like we had got some questions about like, oh, well, if you're injecting routes into the kernel for uh double zero network, like aren't you just monkeying with like the routes in the kernel table? I'm like, what do you think your ISP does right now? Like you're getting all those routes from somewhere. The computer doesn't magically know where every computer on the internet is. Um, but sometimes what we're doing is we're shining light on components of blockchain that are today highly centralized and no one thinks about as risk factors. And then when we talk about them, they're like, oh, I didn't realize like how centralized this component actually was. But there's this kind of period sometimes they'll go through of thinking we are a centralizing problem. We're like, no, no, no, you just have never understood the risk profile you've always operated on. That's always an interesting one.

SPEAKER_00 1:11:06

Okay. Yeah, I appreciate that. Um curious. Uh, what is your North Star metric? And then what is, and I think I talked about that with the validators, but just what is your North Star metric? And then what is one key metric that you know the audience should pay attention to and maybe check on quarterly? Um yeah, and is there alignment between the two or are they different because your different like roles?

SPEAKER_01 1:11:29

I think they're they're probably there there's like important specifics for a network like Solana, and then there are like much more you know generic double zero growth type activities that are are going on as well. You know, at the end of the day, blockchain is still not using a ton of data. Like Solana is like an incredibly fast blockchain, and it's not really still using that much data on a global scale of things. And so um, what I'm hoping is we'll see an era where developers are saying, we're gonna see what happens if we use a hundred times more data and set our protocol limits a hundred times higher than they are currently. For me, that's gonna be the really interesting kind of metric. But, you know, this is why we need a uh a whole diverse set of use cases in and outside of blockchain, because that sort of provides some base load capacity to help encourage more contributors to come online, connect more data centers, and sort of scale up the speed and performance of the whole system.

SPEAKER_00 1:12:26

Very cool. Um, well, well, speaking of you know, a diverse set of uh use cases and stuff, if if we could just go back to that big vision um and and kind of to leave people with this uh to give them hope and optimism on the on the way forward. Um if you could fast forward five years, what does a world with widespread double zero adoption look like?

SPEAKER_01 1:12:46

So you might think there's a way to pay for prioritized data over the public internet, but there's simply not. Like if you are Epic Games and you want people in third world countries to have good experiences playing Fortnite, you can't just say, hey, this is class two data. We're gonna pay more money for it than everyone else. There's no universal flag or system at all. So instead, they have an army of folks that go around and kind of just like bribe local ISPs in like a lot of these countries to be like, look, we we really want kids to be able to play Fortnite because then they'll make more money selling skins and you know, all this stuff. And and and um you that is not a system that exists on the public internet at all today. So one of our goals, I'd say there's kind of two components to this. The first is you're gonna be able to go get a high quality, high bandwidth, high availability, low jitter, low latency connection anywhere in the world as easily as you can provision a server on the cloud. Like today, if you say, hey, I want 10 gigabits of connectivity between Chicago and Tokyo, it's a months-long process to actually get that connectivity. It's and so it and like you probably will have to send a fax at some point. Um, and so the beauty of the cloud is you can show up with a credit card and 10 minutes later you can have 45 servers up and running. And, you know, we hope to bring that same kind of turnkey experience for high performance networking. Um, we also think there's a ton of applications for anytime someone wants to prioritize any arbitrary type of traffic, they're gonna be able to do that through the double zero protocol. And that's you know, a five to 10 year vision. We're we're certainly nowhere near there today. Um, but that's the type of thing we're thinking about is is truly building a parallel internet for high performance systems.

SPEAKER_00 1:14:34

Dude, I love it. And about over a year ago, you know, I was telling people that the world, and especially in our world, it's it's bifurcating, right? You've got fundamentals, it's like this barbell, fundamentals on the right, memes on the left, and this is as fundamental as you get. Um, and and that that's something I can dig in. Um, so I want to I want to thank you for coming, Austin. It was it was a pleasure to have you. Um, and just to summarize for everyone, you know, today we covered the internet bottleneck. We we talked about how double zeros underlay works, we discussed some early metrics, um, we discussed you know the incentives that govern the system and and uh the decentralization of the system um and how things will be open and fair. So, Austin, I want to thank you again for joining. Thank you.