High Speed Horizons Speed and Scalability Converge

So I want I want to start with something I heard you say in another interview because I was watching some of your uh interviews in preparation for this where you said that u the idea behind Salana was a engineering idea not a marketing idea. So take us through that origin philosophy if you if you can.

Um so I'm I'm an engineer by trade. I spend most of my career at Qualcomm working on operating systems uh protocols you know optimizing things making things faster. And around 2017 is when smart contracts I think really came to the forefront. And because I spend most of my time in embedded systems, there's actually a lot of parallels looking at Ethereum at the time and how much how resource constraint that system was. You had this great new platform for building financial applications that could actually minimize risk in software to the point that you can trust it with money.

That that was a a brand new thing that never existed before, but it was very uh relatively slow and constrained. you know, you can do seven to 11 transactions per second. And I, you know, as the legend goes, this actually did happen. I was at Cafe Sole in San Francisco at two coffees and a beer. And that didn't sit well with me.

I was up till 4 in the morning and had this Eureka moment. Uh, applying a lot of parallels from TDMA, which is kind of one of one of the first cellular protocols, one of the first optimizations that folks did with radio to blockchain. and my back of the envelope uh kind of calculations where it's you know a thousand times faster and that's really what uh pushed me out the door uh to go build Salana but um making something faster doesn't mean that you know what it what it's for what it's going to be used for and we kind of instinctively called it blockchain and NASDAQ speed because we thought a new use case that maybe this performance opens up is is trading and being able to globally guarantee that you have the same markets and the same market information available everywhere. Um that everyone can access them permissionlessly. Everyone can create markets and and trade them.

And um you know like all these things actually came true in retrospect but the path that we took was really unexpected. Um you know none of us could foresee that NFTTS would be a thing or or things like that. I want to stick with things I've heard you say because I think these are very salient about Salana and your personal philosophy. You also said that in the process of building Salana, you weren't trying to reinvent the wheel that you were looking for existing computer science, existing computer software uh design to build Salana. Take us through that and take us through why you made that decision.

Um, a lot of our advantages came from I think because myself and most of the team were very experienced engineers. Most of us had more than 10 plus years building software at scale. Um and we were competing actually with a lot of teams that were I think more academic and kind of looked came out of universities and um we took an approach where we used as much offthe-shelf solutions that we could um including like you know for example the virtual machine actually came from the Linux kernel. It's the the BPF VM which is proven at scale to be high performance. Um, so just those little decisions I think was why we were able to ship uh much faster with a smaller team.

We we didn't have to raise as much and um being first to market with a new thing I think is worth a lot. Like I think if you talk to startups and you kind of see which startups survive and which don't. Um you know you can look at even like OpenAI is I think is the most recent great example. they were first to market with a GPT implementation and even if others catch up, it's very hard to dislodge the the market leader. Um, and we were, I think, the first high performance chain to launch.

We launched what we thought was almost the worst time in the world. This was March 2020. Um, right after kind of the double black swan, the Bitcoin and S&P 500 crashed like 60%. NBA was cancelled, so I couldn't watch the Warriors. Um, but we were done and we could launch the product and we felt that you can never time the market, but uh the worst time to launch is right before this kind of crash.

So given that it just happened, it seemed like maybe it is the best time. And in retrospect, um, a lot of our competitors kind of froze up, I think, or or maybe hung hung back and took a little longer to to see how the markets played out. And that gave us that three-month window to establish Salana as a brand for high performance uh, blockchain applications.

It's such an interesting point from an entrepreneur's perspective because you you you know, Reed Hoffman once said to me, if you're waiting for the website to be perfect, then you've waited too long to launch. So you guys got out there were first mover. Uh like everything in life there were some glitches and some outages. So So take us through your management and leadership style going through some of those speed bumps that you experienced and what did you learn from that and talk about the stability of the network after that?

Oh for sure. So um Elon is one of my favorite engineers and I mean his philosophy is uh you ship and iterate as fast as you can and we were always paranoid about safety uh and high performance and those were the two kind of most important variables that we were optimizing for. So if we had uh downtime it was because we took a lot of um engineering risk to both optimize for both safety and high performance at the same time. and you fix those bugs and iterate and you kind of um build a lot of shared trauma with the engineers. This and this actually kind of carries you through through a lot of future challenges because those folks that work together, work closely together through those like nights uh and weekends actually end up gaining senior seniority much much faster than you would in a any other traditional environment.

talk about the stability of the network today and the throughput of the network. Make a comparison for when the network was experiencing outages and where the network is today. I mean, you're probably six or eightfold in terms of the throughput through the uh

Yep. So, um I think it's been nearly two years since we had any uh performance degregation or downtime. So, 100% uptime for almost two years. and capa block capacity is I think is almost doubled. So there there's now twice as much capacity for transactions.

I think on any given day Salana does more transactions than if you look at the top 50 blockchains combined like look at coin market cap and add up the TPS of all those. Um Salana usually I think does more than all those combined. And the other part of it is kind of our our um most famous ethos is IBRL increase bandwidth reduce latency. For finance especially trading the latency part is is as critical as bandwidth. So reducing latency is you know one of the things we're constantly optimizing.

Um the confirmation times are you know sub 600 micro microsconds at the medium milliseconds sorry.

So how how should we think of Salana? Is it is it is it iOS? Is it uh the New York Stock Exchange?

I think

is it the phone system or what how would you describe Salana to

I think it's kind of um I think of it as more the if New York Stock Exchange is Microsoft Salana is the Linux version of that because I think for um finance is a is a is regulated. It's uh a very kind of system that every sovereign state doesn't want to give up control over. So it's a even if you build a technology that can do finance, you're still have um to to deal with like integrating it with the financial system as it is on the ground in whatever jurisdiction you're in. And if you have an open, transparent, permissionless system that's open source, um that means nobody owns it. So the folks at France in France running Salana nodes have the exact same control over Salana as somebody in United States, somebody in Abujabi and because of that they can actually use it as this trusted uh neutral layer that they can all communicate and trust each other and you can't do that with the New York Stock Exchange that's based in United States.

that technology is sovereign effectively under the regulators in control in United States and that may be a fantastic great product but you cannot easily export it everywhere in the world at the same time. Um so just like kind of Linux opened up operating systems and gave them to everybody in the world. Um our goal is to do that for finance for high performance finance. Um, and I think Salana is, you know, the furthest ahead of all the other folks trying to trying to build this and achieved, I think, the most in practice.

So, I I if you don't mind, I want to get I want to have a little bit of a speed analogy comparison. So, uh, the NASDAQ speeds, Visa, Mastercard speeds, New York Stock Exchange speeds, how does Salana compare?

Um, so there's a bunch of different ways to measure it. One way you can measure it is actually settle trades per day or something like that because a lot of the protocols that are built in the New York Stock Exchange are actually very regulated. It's uh the central limit order book and how market makers communicate it all the all those things are effectively fixed by regulators. You cannot innovate there. The advantage that Salana has is that um it's open and transparent and permissionless and that means developers can innovate much much faster in that environment and the protocols that they've built are just much more efficient.

So the result is that with less bits, fewer messages, fewer transactions, you can achieve the same amount of actual executed trades per day as New York Stock Exchange or NASDAQ. I think um depending on market conditions, I think it's hit like maybe 50% of of actual trades settled um of the of NASDAQ.

You know, when I when I think about where you are and I think about the regulators, it's a little bit like Field of Dreams. It's like you built it and now we want the regulators to come. You know, they say we'll build it and then they will come. Where do you think global regulation is? What do you think US regulation is right now?

I think um the slowest part in kind of blockchain adoption is the the human brain almost like um and you know I grew up with the internet in the '9s and as a kid I was just immediately adapted to it like I immediately got it but you know my parents didn't and it took him a long time to understand what a link was like to build that mental model of clicking on a link that it takes you to different document so that was I think The slowest part of internet adoption is just the human brain adapting to this kind of new data structure that's common. And blockchain is a new data structure. It's a common ledger that represents finance. And it has cryptography primitives that are hard to understand. And it just takes time to adapt not just for my parents but for regulators as well.

And I think as the industry matures and it shows that it can deliver finance cheaper and faster to consumers with reduced risk of you of fraud and reduce risk of defaults and actually show that in practice which I think it has you know repeatedly over the over the last few years the regulators want that they actually like in their that's their job and they're all actually want to improve the markets that they regulate and they will adapt this technology, but first they have to understand it and then they have to see that the theory actually matches what happens in the real world and what what matches in practice

because you you you talked about March of 2020 which was a really rough time for most of us in the financial services industry but I want to talk about March of 2030. So it'll be 10 years after the launch. Where are you? What are you dreaming about when you're manifesting the future for Salana? Where would you like to see it be?

Um, I mean, I'm gonna kind of echo a lot of what the folks said right before this panel. Um, I think the world is, you know, something like 500 trillion worth of assets altogether. And if you even look at

is Elon going to own half of that or no? Right. No, maybe possible.

That's the goal, right? Um, I think the blockchains that survive, the open public blockchains that survive are the ones that will try to tokenize all of it. U, that doesn't mean that one will actually succeed at that, but I think the only ones that will end end up existing in by 2030 are the ones that are trying to do that. And if you look at the entire sum of all real world assets and stable coins and everything put together, I think we're at like half a trillion right now. So, this is what.

1% of the way there. So I think um even at even by 2030 if we're at like 10% of the world's you know 10 20% of the world's assets already being tokenized that would be a very dramatic change to to how finance is done. Um, and it's hard for me to actually grasp all the, you know, side effects of that, but you know, obviously what I really want to see is faster, cheaper for consumers, better price discovery for consumers, less fraud, and less risk.

Is there things that Cholana will need to do over the next five years in terms of its software innovation, software adaptation? What are some of the things you're thinking about there? Um, a lot of this is again IBRL, increase bandwidth, reduce latency, and that's kind of our day-to-day motto. Every sprint there is an improvement in one of those parameters. Um, so I'm actually very confident that the, you know, any kind of hardware or throughput or latency kind of potential speed bumps or issues by 2030 are going to be solved.

But just like any technology, it's never really done. So every two years, you get an improvement in hardware from TSMC, better process, better chips. You'll always have more work. So we're never going to, you know, there's always going to be engineers improving it. But none of the problems that are going to arise from scaling to support all of the world's finance or, you know, supporting all of the world's assets uh are insurmountable.

like all all of the stuff in my mind is already kind of um downhill as far as engineering is concerned.

You know, one of the things I've always admired about you is that you're not in competition with anybody. You're just extremely focused on the engineering process and so forth, but if you were to make a comparison to other chains and you don't necessarily have to name them, what was the competitive advantage that you were looking for? Speed is one of them certainly, but there's also usability, right? ease of use, right?

Um, I am actually very competitive. I love sports. So, I kind of look at any transaction that exists somewhere else should be running in Salana instead. So, we're all competing. Uh, and that's good.

I think actually the industry improves from competition. It's the best way to improve technologies when we can all compete. Um,

well, you're a gentleman about it. I never see you like going after anybody, but

um, and I want to win. I like winning. I think um I I think the the the key the key parts that are we we unlocked for usability were actually pretty simple. It is just speed and cost to users. I think those two are so important that if you have uh friction in terms of cost and even if it's intermittent friction like once in a while the fees spike or something like that that really could break a lot of business models that could break a lot of users.

Um, if you're you're a merchant and you're selling coffee, you can't have your onchain fees, like your processing fees spike to a dollar all of a sudden, even if that is, you know, is intermittent because all of a sudden your business is effectively down for that day. Um so from our perspective guaranteeing a quality of service in terms of how fast transaction execute and how cheap they can execute those two are kind of the key most important parts for for UX. Well such a great message for everybody. Listen I'm a as you know I'm a very big bull on Salana. We own a lot of Salana at Skybridge and congratulations on your success and I'm looking forward to being a uh a long-term holder of Salana and I'm looking forward to sitting down with you in March of 2030

to discuss the uh the great things that you're doing. Ladies and gentlemen, thank you very much.