Supercharging DeFi and Social AI Systems by Creating the Most Democratized Network, with Steven Pu @ Taraxa (Audio)

[00:00:03] Hello everybody and welcome to the Crypto Hipster Podcast. This is your host, Jamil Hasan, the Crypto Hipster. We're interview founders, entrepreneurs, executives, thought leaders, you name it, all around the world of crypto and blockchain. And today I have another amazing guest. He is the co-founder of Taraxa. His name is Steven Pu. Steven, welcome to the show. Thank you very much for having me, Jamil. Great to be on the show.

[00:00:30] You're very welcome. Very welcome. So I'll ask you first, and I'll ask you, what is your background and is it a logical background for what you're doing now? Yeah, so I suppose it's somewhat logical. I was educated as an electrical engineer at Stanford University, a bachelor's and master's degree.

[00:00:54] So I do have a technical background, but my career hasn't been as an engineer. After I graduated, I did consulting and then I did a bunch of startups, but I wasn't sort of the primary engineer in those startups. So I had a lot of startups. So I had a lot of startups. So I had a lot of startups. So I had a lot of startups.

[00:01:14] And my career had been more in business. I taught myself how to code when I was like 12 and happened really just kind of coding random stuff on my own for fun. I did a bunch of startups when I did a bunch of startups when I was actually in high school during the internet boom, actually, you know, during the late 90s. So I've always had a love for engineering. I've always had a love for building things.

[00:01:41] I did multiple startups when, you know, prior to Tarexa in Internet of Things and mobile health. So not really a stranger to entrepreneurship, not a stranger to engineering. I just I just happen not to have an academic background in things like distributed systems, for example. So I had to read a lot of academic papers when I first got into layer one design.

[00:02:10] Got it. You know, I was I want to find out what block I want to find out what Tarexa all is all about, including including what a block dag is compared to a block chain. Yeah. Yeah. So Tarexa is interesting. We started thinking about. Kind of figuring out what we should do in Web3 once once I got convinced that decentralization is truly something extraordinary and that I wanted to be a part of it.

[00:02:40] So that was back around like early 2017, late 2016. So a little bit late to the game. I first heard about Bitcoin in 2013 took me a few years to really to really catch on. And once we figured that out, you know, we're like, what can we do? Right. What is the most interesting and impactful thing we can do to kind of get into the space?

[00:03:04] And layer one was at the time a pretty obvious choice, because as we all know, Ethereum just came out and smart contracts just kind of came out. ERC 20s came out, had a big ICO boom. And we all saw that the state of the art, right, at the time Ethereum was not really that scalable. And for any kind of meaningful applications you run on top of a decentralized network, you need much better scalability.

[00:03:32] So that was that was really the entry point of how we kind of decided to build layer one for Tarexa. And at the time, we were searching for different types of architectures. So what kind of architecture would really work well for something that was scalable, but at the same time didn't sacrifice any decentralization or security? Right. So Vitalik had that classic trilemma that he posed.

[00:03:58] So we said, you know, this is often the problem that the first few iterations of a technology tend to be suboptimal. Right. You see a lot of tradeoffs, but really the tradeoffs exist because you're at a suboptimal point. So it seems like whenever you do anything, you're sacrificing something else. And those tradeoffs in engineering, you always get tradeoffs. But because you're stuck at a suboptimal point, you are trading. You can't really improve.

[00:04:27] So you need to kind of shift the paradigm and move it to a different place where everything just improves without trading off because you're no longer stuck in a suboptimal location, overall architectural design. So that was what we're looking for. And again, we read a lot of papers. And one of the papers we actually came across was this paper called Phantom Gold Stack. It was authored by Son Holinsky, Wyborgski and Zohar.

[00:04:53] So for people who know, Son Holinsky and Wyborgski are two of the key founding members of the CASPer project. So these are the people that actually invented the blockback paradigm. And they have been, especially I think Son Holinsky, have been publishing very academic and well-researched academic papers with security proofs for many, many years.

[00:05:18] Like after Bitcoin came out, there wasn't really a lot of mainstream academic interest in this technology. And he was one of the few serious researchers that was actually looking at it. So he is very well known within the blockchain, the crypto circle for the rigorous and just really high quality research. So we saw that paper. We read a lot of papers which had a lot of different things. We saw that paper and it was like, wow, this makes a lot of sense, right?

[00:05:46] So block DAG, as it was described in that Phantom Gold Stack paper, resides on a very, very simple concept, which is instead of just having one block added to the network at a single time, why can't we just add more blocks? Because the network actually naturally on single chain architectures, most of what you see today are single chain architectures, which is like Ethereum, like Bitcoin. They add one block to the system at a time.

[00:06:15] But in the network, more than one block is produced at a time. And all the other blocks that are going to produce are thrown away. So this paradigm, this block DAG paradigm just says, look, why don't we figure out a way to include all the blocks that are produced and we don't throw them away, right? Because there are many nodes on the network. Why not take advantage of the natural parallel processing that just exists naturally on a decentralized network?

[00:06:42] So the difficulty then becomes, well, if you include all these blocks, how do you ensure that ordering is secure, that you don't get reorganized, you don't get attacked, and that the network still remains decentralized and secure? So this was the big problem I think block DAG solved. They had a few kind of very interesting examples.

[00:07:08] So in a single chain network, how do you increase throughput, right? So you can only add one block at a time into the network. And so if you keep that going, if you want to remain one block at a time, what you can do is maybe you can make the blocks really big, right? Just make the blocks bigger. Then maybe you'll get more throughput. Maybe you scale up.

[00:07:33] But the problem with much bigger blocks is that they propagate through the network much more slowly. So that creates a problem in which one node is much more likely to see a different picture of the network than another node. And when different nodes see different pictures of the network, they start building on different blocks. So they start creating these forks on the network.

[00:07:57] So not only are you not really increasing the throughput, you're also sacrificing a lot on the security because now you get a lot more forks. You get a much higher risk of reorganization and rolling back transactions. So increasing the size of the block doesn't really make all the sense because you're massively trading off security. Okay, so why don't you – okay, so increasing the size of blocks doesn't really work. So why don't you just create a lot more blocks?

[00:08:26] Well, if you create a lot more blocks, you run into the exact same problem because at any given time, only one block can get added to the chain. So if you create a lot more blocks, then you just end up confusing more nodes, right? And you get the exact same problem where all these nodes are seeing different blocks, lots of different options at any given moment. And they're adding different blocks into the chain, creates more forking, sacrifices more security.

[00:08:51] So these are fundamental problems and challenges that single chain architectures run into. And it's extremely difficult to overcome them unless you change the fundamental architecture, which is turn this into – from a single chain into more of a graph, right? You allow all the blocks to come in, be ordered according to a specific ordering mechanism. They actually describe it in the paper called the GOES protocol.

[00:09:18] And that ensures ordering and ensures that there's not a lot of reordering going on. So when we read that paper, I think they just published it. And we said, wow, this is pretty amazing, right? And it looks like it just makes a lot of intuitive sense. They also have very rigorous security proofs. But intuitively, it just made a lot of sense. And there are a few open problems back in 2018 when we first saw the paper.

[00:09:46] And so we decided to say, hey, let's try to adapt this topology, this architecture, into building really the next generation of layer one architectures. And at the beginning, when we started building this, the paper described a proof of work kind of approach because it was trying to improve Bitcoin, right? So for us, we said, look, we wanted to build something that was more of an application platform or smart contract platform like Ethereum.

[00:10:17] So that's where kind of our approaches diverged, I guess, from Casp. At the time, it wasn't known as Casp. But I think they, at the time, they were known as DAG Labs. So anyway, a lot of name changes. So we ended up building something that was intended to be a smart contract platform that runs applications. So from day one, we said it's going to have a virtual machine.

[00:10:42] It's going to be 100% EVM compatible just because the ecosystem out there, all the builders out there, the tooling out there are all EVM compatible. So if you made it EVM compatible, then builders will have a much easier time adopting your technology. They don't have to learn anything new. They just use the same tools, and it would just work. So it had to be EVM compatible. We wanted to be proof of stake just because we thought that proof of work has this role.

[00:11:11] I'm not here to kind of slam proof of work. But it does create a ton of energy consumption because of the natural arms race nature that of the hardware arms race nature that creates proof of work. So we said, look, I mean, if you want something that's more efficient, then you have to go with proof of stake. So this is why we ended up creating something that is a block DAG,

[00:11:38] but it's a proof of stake and also 100% EVM compatible. So this is Tracet now, right now is the world's only proof of stake block DAG and 100% EVM compatible smart contract platform. And it's been quite a wild journey because we had to create a lot of innovations that didn't exist anywhere in the paper or anywhere else, actually, in order to make sure that we ensure rapid convergence in the DAG.

[00:12:07] We made sure we were actually able to natively integrate a VM onto the block DAG while removing all the throughput bottlenecks. We had to add another layer of consensus on top of it to make sure that the security guarantees are not broken while at the same time we can maintain high throughput. So we added a lot of different, we had to invent a lot of stuff in order to make sure that, you know, Terraxa as exists today worked as intended

[00:12:35] without making any crazy decentralization or security sacrifices. Yeah, so it's been, like I said, it's been a wild ride. We started in 2017, 2018. We launched our main net in April of 2023. It's been running ever since. So you had a six-year, I guess, build process? Yeah, we had a really long build process. The interesting thing is at the time we were in touch with,

[00:13:03] obviously we were talking to a lot of other teams or building layer ones. And one of the interesting things that was different from what we did versus what they did is that these other teams were much more, I guess, market-minded, that they prioritized releasing much earlier. So we also, and then when they released an earlier product, it was kind of half broken or it missed, it didn't have like 80% of the features that it promised,

[00:13:32] but they got to market earlier and they just kind of kept building and kept fixing it as they went along. And we saw a lot of teams run into a lot of problems because, you know, it didn't take enough time. And they recognize that. It's not like they're trying to trick anyone. You know, you make the trade-off, right? So we always thought that it's for layer one, you shouldn't rush it into the market

[00:13:59] because there's like people's money on top of it, right? In the Silicon Valley ethos, you move fast and break things. If you move fast and break things on infrastructure technologies like layer ones, you start causing a lot of damage. So we had a small team. A lot of other layer ones also had small teams. But I think what made us different than them is we kind of just wanted to release something that was fully functional, fully featured.

[00:14:28] All, everything that we set out to do in the white paper, we wanted to deliver. And we just released something that just worked really well. And yeah, so I think that delayed us for, yeah, for at least several years. There was another reason why I think we were a little bit delayed. We had a fully functional product, a network by early 2022, which I think it's the same time I think Casper launched.

[00:14:58] But when we tried to say we want to do like a native token conversion, so kind of convert the ERC20 into a native coin, the exchanges that we were listed on said that they wouldn't support the conversion early because there's a long queue of people. You have to stand in line and wait for the conversion. And the wait time was like a year. So we were a little bit unhappy about that. But if you did a native token conversion and the exchanges didn't support it,

[00:15:28] that's a huge problem. So we kind of just, you know, we sat around, we did like another audit, we did like a second audit, we added more features and did more testing throughout that year. And we launched, you know, without incident in 23and3. But, you know, if we didn't have to wait for the exchanges, we would launch, I think, early 2022. If we didn't insist on delivering the fully featured version,

[00:15:58] I think we probably could have launched in 2020 maybe. So anyway, so just again, different trade-offs, different decisions. We're ultimately, I think we're pretty happy about the product that we've released, obviously open source, because we still fundamentally believe that, you know, layer one needs to be stable, needs to be secure, and it just can't have any problems when you release it, right?

[00:16:26] So that kind of trade-off was something we just didn't, we weren't waiting to make. So I just got a fresh release that popped up my desk 10 minutes ago, and it was on Blockdag. Really? From whom? I'm looking at it. It says Solana TVL and AFOS price surge as Blockdag fuels DeFi boom. I don't know if that's you that's going to be listed on all these exchanges,

[00:16:55] but I want to find out what the, now, you know, what are the factors influencing your resurgence, and then how, what is your potential to reshape the nature of, you know, whole entire Web3 movement? Yeah, yeah. I think Blockdag, as usual, you know, kind of price action seems to drive attention, right? So as, as, as, as a builder, I rather not, wasn't the case,

[00:17:25] but that's the case. That's how the market works. And, um, I think the reason why Blockdag gained a lot of attention was because Casper had a, just an absolutely meteoric rise in its valuation from the time it launched in 2022, basically getting, it's worth nothing, worth zero. zero. And then, till today, I don't know what the valuation is, like five billion or something like that. So it, it would just, went asterisk, like sort of through the roof. And, um, you know,

[00:17:55] actually this is very interesting. I, I always thought that people didn't care about tech, right? Um, and maybe they don't, maybe they only care about tech because of the price. I'm not, I'm not really sure. The vast majority of people in crypto don't care or understand the technology. So, at Taraxa, we didn't really spend a lot of time talking about technology. I think very early on, 2017, 2018, parts of 2019, we did talk quite a bit about the underlying

[00:18:24] consensus technology, but very quickly, we tried to pivot into use cases. And at the time, we were very focused on enterprise use cases as most of Web3 were at that time before DeFi came out. So, we were always very focused on talking about how this could be used, all the use cases that could be possible, and I think when Casper came out, and then we became aware of Casper, I think, maybe June or July of 2023,

[00:18:53] and there's a huge amount of interest in blockback technology. So, that was kind of coincidence because, honestly, I've been kind of trying to keep tabs on Dag Labs and then, you know, that company kind of disappeared. so, so I figured it'd die. You know, I didn't know and I had no idea that, you know, Casper got launched from Dag Labs. So, I thought that project died, but, you know, so I sort of lost track of it. And, and now there's like kind of a huge amount

[00:19:23] of interest in blockback. and we've had some conversations with some of the founders over there at Casper because we were, I feel very fortunate to be able to talk to them because these, these are the authors of those, the original paper, right? So, it was really, really cool to be able to talk to them. So, so there was a lot of interest, I think, in blockback because of the, originally, I think, because of the price action. I mean, it's fine just because blockback is actually superior technology. Like, we've always said that it's the next iteration

[00:19:53] in blockchain or in decentralized networks, right? And, it's just a natural evolution, right? You go from single chain to like a graph. It's a superset. It's not, it's not extremely accurate but it's sort of like a more generalized version of a blockchain. So, it's, it's just, it's just better technology in every, in every single way. And, how I would see it, I, you know, you already start seeing a lot of projects kind of very quietly

[00:20:23] starting to, adopt blockback like architectures. And, I think today, these days, projects don't spend a lot of time writing out their technical papers. So, it's hard to kind of tease out exactly what the architecture is and we don't have time to like read through the code of all these new projects that are coming out. But, it's pretty clear that there, you know, there's recognition that this type of DAC topology is the way to go. So, I think some of the most dominant

[00:20:52] single chain topologies today, quite frankly, I think they're going to get phased out in a few, maybe by the next cycle they're all going to get phased out. And, they're just, they're just suboptimal. Again, you know, single chain topologies are suboptimal. They're just not, they're just not that great, they're just not that great, right? So, I think DAC topologies, block DAC specifically, would be making quite a bit of impact within them, within Web3 as more and more layer-wide infrastructure

[00:21:22] start adopting either block DAC or something similar to block DAC. As of today, I don't think anyone, any other project really adheres or officially adheres to the block DAC paradigm like Terax. I mean, obviously, Casper, right there, the ones who invented it, but really, we're the only other project I think in the world that adheres this closely to the original paradigm. Other people, like I said, it's unclear, you know, whether or not they just have their own version

[00:21:51] or whatever. But, but yeah, it's, it's, it's, I can definitely see it influencing other projects, which is a great sign, right? We're in this space, everything we build is open source. You know, we hope that the things that we build will have a lasting positive impact, you know, to not just ourselves, but to the entire space. Yeah. Got it. So let me say, make sure I can understand this correctly. And then I have a couple other questions too. Sure. In the past,

[00:22:21] I've interviewed Hedera and the graph, they're a graph, right? And then I interviewed, it was Internet of Things project back in 2021 one or two that I, that I interviewed after, a popular one. They, you know, they had, they had a DAG architecture. How do you differ from the, from the graph and the IoT DAG? Yeah. So, so DAG is a mathematical construct, right? It's, you know,

[00:22:51] it's, it's not like a, it's, it's like a data structure in math, right? It's not, you know, just saying your DAG, it's, it's kind of saying I'm making a round cup. Well, you know, lots of cups are round. I mean, you don't, you don't own the circle, right? I mean, no one owns, no one invented the circle. It's just like discovered it's mathematics. So DAG is like that. It's just, it's a mathematical concept. It's a mathematical construct. And a lot of data could be structured

[00:23:19] in a DAG format. So, block DAG is a specific way of organizing blocks, a specific way to drive convergence in ordering. So, it's, it's very, very specific. I think, I think just because a project says it's DAG doesn't mean it's block DAG, right? So, they might, maybe they could have named it differently. They could have called it like something else that doesn't have the word DAG in it. Maybe then it'll, it'll be a little bit

[00:23:49] more differentiated. Yeah. So, plenty of projects use some sort of graph in their, in their data structure in the way they organize their data on chain. But, it really, I haven't really seen anyone besides Casper, obviously, that does it exactly like the way the block DAG describes it. obviously, you know, everyone has trade-offs, you know, the way this project does it is different than the way the other project does it.

[00:24:19] You know, I think that's probably a little bit too much to dig into during this interview. But, but yeah, just because someone's DAG doesn't mean it's block DAG, right? So, got it. Got it. So, going back to 2017 and 18, I think it was more 17. Yeah. We, we saw a war between people who were big blockers and small blockers, right? We saw that in the Bitcoin that needed,

[00:24:48] that maybe they really needed a hard fork but they ended up with a soft fork and created Bitcoin cash or SV, like, and you were around that time. So, what did you learn from that war between the Bitcoin, the big blockers and the small blockers and the Bitcoin wars and how did you, how did you make a difference in what you built? Well, several things. I think from a high level you start realizing that single chain architectures are kind of not going to be the future, right? Because

[00:25:20] the stuff that were being proposed to scale Bitcoin were just kind of very small incremental steps that had clear trade-offs, right? So, if you wanted to, again, jump out of that, you know, just drastically improve it, you have to jump out of that paradigm. Now, Bitcoin is revolutionary because it was the first one. It's just, there was nothing, I mean, there were one or two things before that might have been similar

[00:25:48] but it's truly revolutionary because it's really the first mass adopted, you know, decentralized network and the person, whoever, he is, right? Designed it and built it obviously went with the most conservative design which makes sense, right? Like, when technology comes out they tend to use conservative designs to make sure that it works and is secure, right? Now, so that's the first thing that we really learn is that this architecture, this topology of single chain

[00:26:19] really is not the future. So, you got to start looking for something drastically different, right? In order to scale. The second thing you kind of learn is that and this happens I think in every single ecosystem is that once they become big enough they start to slow down in the way that they change and it's natural, right? a lot more people are in it, a lot more financial interests are in it, a lot more stakeholders naturally things slow down. It's like a company, a small startup

[00:26:48] moves a lot quicker than a large corporation, right? It's the same idea. So, that part of it is a little bit more difficult, right? It's a little bit difficult to solve, right? Because you want these networks to be decentralized. You don't want there to be like a dictator at the top saying we're doing this today, we're doing that tomorrow. That's not how it really works. But, the more it decentralized it becomes, the more successful it becomes,

[00:27:17] kind of the slower it innovates. I don't really know that there is a solution to that. It's human behavior, right? It's not a technological problem, it's a human problem and it hasn't really been solved over thousands of years of civilization so I'm not really sure it's going to get solved here. The only way that I think markets solve this problem has been the same way they've solved this problem for God knows how long is that the poorly performing networks, the ones that

[00:27:46] don't adapt fast enough, they die, right? And the new ones kind of try to take their place. And we see that happening all the time, right? In the market today we see that happening. The networks that don't innovate, that don't catch up, they don't get as much traction and after a few cycles they kind of just lay by the wayside and people kind of forget who they are. And that's just the way networks, or that's the way everything really works. And it's being played out obviously in a decentralized world as well. So I think those are the two

[00:28:16] things that we learn I guess. But the second one, not much you can do about the first one, it really drives you to look for the next iteration, the next new thing that's really going to happen. Maybe there's a way to foster a kind of aggressively growth-minded ecosystem perhaps, I'm not really sure. But again, you know, some of

[00:28:45] the more successful entrants after Ethereum have also beginning to show signs of ossification, right? They're not really changing, they're not really catching up. Not going to name any names, but just to say that what is winning in this cycle very well might not be winning the next cycle, just because we haven't seen them actually innovate, catching up to the latest innovations.

[00:29:16] It's very paradoxical. These projects have so much money, resources, yet they don't really change, right? It's very, very paradoxical. But I mean, that's just the way it is. The new comes in, the old moves out. Let's talk about the new, right? Let's talk about Block Dag, and you conquer the scalability trial, MS. I want to find out what that is. And then how do you do that? And then

[00:29:45] how do you help DeFi or go on to help DeFi meet its original promises? Sorry, what was that last part? How do you help DeFi meet its original promises? Okay, yeah. So how does this solve the trilemma? So I think at a high level, as I mentioned earlier, it's really at a very high abstract level, it's really about if you're drawing a curve, right? And you're saying, well, where's the maximum? But you're kind of stuck

[00:30:15] in like a little hill on the side, right? If you're stuck on the little hill, it seems like wherever you're going, it's going downward. So you're like, we're at the maximum, but then there's a much higher hill somewhere else. So single chain is kind of stuck on that little hill and they don't see that there's a higher hill where you don't need to make any trade-offs and just kind of improve everything at the same time. So I think I talked a little bit about the security trade-offs for single chains. If you want to increase

[00:30:44] throughput, increasing bigger blocks or adding more blocks onto a single chain paradigm, just sacrifice the security with little to no gains on throughput. So that's the fundamental limitation of single chain architectures. On a block tag, you're naturally allowing more nodes on the network to contribute multiple blocks at the same time. Or they're asynchronous, not even at the same time, just contribute into the graph.

[00:31:14] Adding more transactions into blocks and adding those blocks into the graph very naturally asynchronously and having everyone contribute simultaneously. Do you want one node to be contributing at any given time? Or do you want 10 nodes to be contributing at any given time? No brainer. Obviously the more the better. With that comes the complexity of how do you keep ordering straight? That's a big

[00:31:43] challenge. With a single chain it's pretty obvious. There's no confusion about ordering. On a graph it gets a little bit harder. How do things get ordered if everything is in a big mesh? I call it a popularity contest. Here's a popularity contest. When you create a block on

[00:32:13] the dag you say when I'm creating this block at the time of me adding this block into the block dag here are all the other blocks I see that are right before me. It's like standing in line. Imagine you're standing in line and who's in front of whom it's not a single line it's a big crowd. All you're doing is everyone said when I got here

[00:32:43] these five guys were in front of me. When I got here these other four guys were in front of me. If you start putting that together after a while you get to see he says these three guys in front of him. I'm not going to go into the math but after a while you can say okay then we can actually start doing some ordering out of this assuming most people are honest

[00:33:13] and they're not lying. So that's how you do ordering that's kind of the ghost rule which is weights it's a system of weights add them together depending on who sees whom in front of them when they're creating those blocks so that's a

[00:33:44] the person you think everyone else thinks is in front of them that kind of thing the metaphor breaks down a little bit when you start getting too deep into it but it basically is who's in front of you and who do you think the most popular guy that's in front of you is right and if everyone does that and you have this idea of building a path that links up the most popular people the most popular blocks in the DAG and that helps you to drive ordering convergence

[00:34:14] across the entire graph so this is a critical problem that had to be solved I think in the original paper for phantom ghost DAG they described a different way of doing it slightly different way to the waiting system the ghost rule is still the same but to drive rapid convergence they had a different way of doing it which actually was very slow but later on as I talked to the Casper team they actually

[00:34:43] it was a really hard problem to solve we actually didn't solve it so we invented a new way to do it and they kept going and I think they solved it and I hadn't had time to dig into their solution but in the end what you're trying to do is you're trying to make sure that there is ordering established and that you want the ordering to not keep changing all the time so that's what we

[00:35:33] so you to something a lot of proof of stake consensus has this issue where they're picking a leader and you know who that leader is so just attack the leader pretty simple so these are issues that they have but this one all the people are randomly producing these blocks at any government you have to attack the whole network right so no decentralization

[00:36:03] sacrifices no security sacrifices but at the same time clearly you get a massive throughput upgrade just because there are a lot of nodes adding blocks into the DAX simultaneously so intuitively this is a way you can start understanding this is kind of lots of people working together and using kind of who's in front of me kind of an analogy to determine ordering and

[00:36:33] it just works really well and it's very elegant that's one of the reasons why we just absolutely love that paper when we read it it's so elegant it's so simple it just makes sense right so that was very the decent decentralized networks

[00:37:03] as much as DeFi so I think there are several things that can really help DeFi number one obviously the scalability so you can actually bring larger portions of the activity on chain right obviously faster cheaper everyone likes that and so it instant

[00:37:33] transaction inclusion right so this is very very interesting so as soon as a transaction gets sent into the network it's almost instantaneously put into a block just because there's so many people see it and one of them is going to put it into a block right so the problem with kind of front running right I think other people call it MEV or there's kind of different names for it but the idea is that you do a DeFi transaction there's a bot kind of watching all these

[00:38:03] transactions they're like wait that one I think I can take advantage of and they try to sandwich attack your transaction making your transaction more expensive they're basically siphoning off you front running lots of different names now the key to front running is that they have to made it into the block yet by the time they've made it into a block it

[00:38:33] just it's too late it's way too late for you to attack it because it's very hard to mess with ordering once it's in the block so because the block that kind of paradigm helps you put these transactions into a block almost instantaneously that means that there's not a big queue where people are waiting in line so right away you're eliminating the

[00:39:05] pool from which the attacker can select victims from they don't even see anything in the pool so what are they going to attack they're not going to attack anything and by the time they attack it's not worth anything right so it also just by this very nature of this block that it can help really to solve this problem of front running

[00:39:48] I is is something that's not really well known not really widely understood but it costs a lot of people a lot of money every single day actually so I think again with a better architecture you're able to make a lot of advancements and simultaneously without making a ton of trade offs this is when you know you have a

[00:40:18] fundamental upgrade right so I this architecture is really going to be a big deal if not this cycle then the next cycle it's going to shine through at some point right sounds like to me I agree I'm glad you're doing the ordering on the technology level because I'll be honest with you there's a lot of four-way stop signs in my town and

[00:40:47] people don't want to go and order so they want to shoot before you but I want to thank you very much for speaking with me today I'm going to follow you guys and probably buy some block dag if I can you know and I'm looking forward to this next year this next cycle so I have one last question how can people find more information about you about Taraxa about block dag how can they do that yeah so first

[00:41:21] interesting information just on the main website if you go to visualizer dot terasa dot io you get a real time visual representation of the entire network which looks really cool I look at it for 30 minutes at a time it is so mesmerizing it very interesting you'll see that it is very different than all these other architectures out there

[00:41:51] you can read all the articles about all the stuff that we just talked about yeah and if you want to find out more about terraxa's ecosystem you go to terraxa.land and all the ecosystem projects there we have DeFi we have Dexas stablecoins you know there's a lot more stuff in the pipeline we have a big social AI ecosystem that's coming out very very soon so yeah give us a follow on twitter at terraxa underscore project so

[00:42:21] awesome thank you very much for your time today thank you very much great to be here