 All right. Thank you, biology for joining us. Super excited to have you on the show today, and we want to talk about decentralization. So maybe you can just kick it off a little bit of a definition and how you know if something's decentralized or not. Sure. Absolutely. Yeah. So, you know, the thing is decentralization, people often talk about it in quasi religious terms. You know, this is decentralized. This is not decentralized. They don't give a precise definition. And, you know, a while back, I wrote an article called quantifying decentralization, which proposed an actual metric. And this metric is we call it the Nakamoto coefficient. It's actually used by Solana, the rising, you know, blockchain as well as other chains. And the concept is, what is the minimum number of entities you need to compromise to compromise the entire system? Okay. So, for example, you have a certain number, let's say it's a coin, okay? You have a certain number of exchanges where that coin is traded. You have a certain number of miners. You have a certain number of developers on the main client. You have a certain number of clients. How many entities, if you shut down three exchanges, would that bring the system to a halt? If you shut down five miners, would that capture more than 51% of the mining power? If you, you know, if there's only one client, could you get three developers and then there were only five developers, you got more than 6% there? Basically, how many entities would you need to capture in order to capture the system? And while an imperfect measure, it does capture the fundamental concept that a decentralization is continuous. It's not just a zero one. You can keep increasing the amount of decentralization where you have to capture 10, 20, 100, 1000 nodes. And be it captures the concept that decentralization is based on the minimum number of entities you need to compromise at any choke point. So you might need to go and compromise 50 exchanges to compromise this coin, but three developers do the trick. So you basically have, you know, like about three people. You're, you know, another way of thinking about it is your bus number is three. And so it's basically a measure, it's like a choke point analysis or a bus analysis, an enumeration of control points. It's how many nodes do you need to take down the system? That is a measure of how decentralized it is. So I think that's a great way of thinking about is something decentralized, is it not decentralized? You talked about coins and chains and that kind of thing. But if we take it a little bit higher, why does it matter if something is centralized or decentralized? What's at stake? You know, what is the kind of ultimate goal of taking this approach to solve some of the current technology problems and challenges that people are applying these tools to? So it's a great question. Decentralization is very computationally costly in the context of blockchains. And, you know, you'll see people say things like, oh, blockchains are just slow databases. What's the point of doing all this stuff? I can just keep it in a database. And, you know, why take all this overhead of decentralizing as you're asking? And the way to think about it is blockchains are massively multi-user databases where every user is a root user. And that is why you decentralize. Now, when we think about this just to be concrete, do you have, you know, the password to your Postgres on the internet? No, I hope not. You don't, right? Exactly. And most of the people watching this would not put their root password to their Postgres, or, you know, it may not be just one database. Of course, you could have multiple databases, multiple secrets, but Twitter doesn't put that password on the internet or those passwords. Facebook doesn't. PayPal doesn't. And so those are what I'd call closed-state services. Their backends are closed-state. You cannot see every row and column in their tables, and you also cannot see how the services are executing. You know, if people accuse them of, you know, this or that algorithmic issue, nobody can diligence that they have to sort of take that on trust and on faith. By contrast with a system like Bitcoin or Ethereum, they're not closed-state. They're open-state systems. It's a next step after open source. It's not just that the source code is public. The entire database is public. And you can look at every single record in the blockchain, every single transaction all the way back to t equals zero, every digital signature. And it's not just open source and open-state. It's also open execution where you can run a node and you can trace every single opcode that is happening and see exactly how the logic of the smart contract is working in practice. So you no longer need to trust the operator of the server-side service. You can run it yourself and you can actually diligence not just the client, which we can do with open source, but now the server, which we can do with these open-state, open execution environments. This is a really interesting progression and way to think about it. As you're talking this through here, I'm thinking through my history in the web world, which is 20-plus years now. And in the beginning, we were building web apps to take data and expose it through browsers and a lot of Internet Explorer back then. And giving people the ability to sort of interact with it through a very tightly controlled user interface. And then one of the things that was really powerful in the next decade after that was kind of the API movement. And a lot of organizations started to create APIs for some section of their data and then allowing people to interact with it. Read from it, learn from it, apply it to other things. In some cases, even push transactions back into it. But I think that the point you're raising here is really important, which is that ultimately there is kind of one entity that owns the canonical backend records there. And you have to trust them to not be modifying them in a way that's invisible and non-public and that you can't see. You have to trust them to maintain that data and not have a corrupted backup and lose it. You have to trust them to not be compromised by very well-funded hackers and people who are out there trying to exploit all of this. And when you move that to an open state, it's similar to open source where you can see not just kind of one particular view, but you can see the entire view and you can see the entire history of how we got to that view. That's a really interesting way of thinking about it. That's right. And one way to think about it also related to that is when it comes to trust, again, people can kind of freak out about this, but trust, it's also continuous. Take it in the context of, let's say, trusting Facebook's APIs or Twitter's APIs or Microsoft or Apple's APIs when you build on top of them. The thing about trust is those companies are powerful companies and they can go up the stack at you and they have. If you get too popular on Facebook or Twitter in the past, they would just kind of look at the app and then they'd just kind of go up the stack and clone that app and bring it into the platform. Microsoft was famous for doing this back in the day. Apple also still does it. And so the use of the API itself is sort of somewhat adversarial. They can also choke off API access to outsiders. Famously, Twitter did this where Twitter initially was essentially a protocol and what they found was that they couldn't monetize that protocol because if they had it fully open, anybody could rebuild the entire Twitter experience and did like TweetDeck did. And then there were no longer Twitter, the company was no longer the monopoly provider of ads. Somebody could rebuild the entire experience and undercut them on the ads. And so they're like, okay, actually monetization requires a monopoly over this user interface. So we're going to choke everybody off. We've actually seen that in practice. And that is because Twitter was basically created before the concepts behind crypto were invented. Today you are and you will see, you're already seeing decentralized Twitters where there is a token that is integrated with the protocol and now the backend is open state and open execution and of course open source. And there's no longer this thing of throttling a client because the developers of those open source, open state, open execution, Twitter crypto versions want everybody to use it because they make money on the coin as opposed to the ad, right? And so what coins do, what tokens do is they sort of abstract out monetization into really just this one unit of this digital asset. And then everything else becomes open. It's kind of the next step after open source where you're generous with your source code and you're monetizing the server side. Now you make even the server a bit open. You allow anybody to run it and you essentially just have this coin and it's effectively the mind share of the community that you're monetizing. All right. So we've been talking about all these different technologies in this kind of web three world where you build a platform and it's very decentralized. So if you build on top of it, you don't have to worry about it being pulled out from under you, the rug, so to speak. But if those are running on one or two clouds, right, then doesn't that also create kind of some risk there? How do you think about the underlying cloud infrastructure options out there and how important is that? Yeah. So the way I kind of think about this is, you know, in 2010 when, you know, serving the space like Google and Facebook and Amazon had these massive data centers. And as a joke went, how many stories does your data center have? And because of this multi-story data centers, you've got like murals on them in the Midwest. You've seen those for Google. And it just seemed like this enormous competitive mode that people couldn't really deal with. And, you know, if you're starting a new company, well, okay, fine, use, you know, one of these services, but you're basically sort of at their whim at some point if they decide to use them adversarily against you, you know, if they promise not to do that. And, you know, this was actually one of the areas where I thought it was going to be the hardest to compete with them and I didn't have an answer for it at that time. But then now, 10 years later, seeing the global build-out of Bitcoin mining and also of Ethereum mining and for the back-ins for Solana and Filecoin and Chia and everything else that requires a hardware build-out of some kind. It's not just pure staking that is compute-intensive or storage-intensive. I realized that actually, yeah, if your function is, you know, like small enough, if it's a mining function, if it's a hash function, or if it's something like storage where it can be exposed also as a function, you can paralyze this massive global data center build-out for these back-ins. And it can happen. It can happen in decentralized ways, many different proprietors. It's actually a very good fit for OpenStack where you can do open mine or open farm and you can help decentralize many of these protocols by providing quasi-turnkey kinds of things, certainly software assistance and possibly software plus hardware so that, you know, people can just sort of turnkey set up a Filecoin mining or Filecoin farming or Bitcoin mining operation somewhere. And I think that is complementary to the existing sort of thing where, you know, certainly people will run nodes locally, you know, in a home internet connection or they'll run them on AWS or they will, you know, run them in a rack in somebody else's data center. But running their own is, of course, the ultimate. And even a few of those worldwide is actually pretty good because you can get tens or dozens of them, maybe even hundreds, depending on how big the thing is. Bitcoin mining, it's probably in the thousands. I actually don't know how many folks are mining right now. It's quite a lot. And so that's, I think, where the opportunity is for OpenStack or the folks in your community is like open mine, open farm and decentralizing the back-ins further. And what that does, by the way, is it shows that Google and Facebook and Amazon, their strongholds are not impregnable. Once you can decentralize the back-end, you can beat them. Thank you. This has been very, very enlightening. Give us a lot to think about. And the last thing I'll just note on that note is that we actually have public clouds powered by OpenStack and over 180 data centers now. We have a map back here. So you have people running it at home, private clouds for big companies, small, but also a very competitive worldwide network just blanketing the world with public clouds from all kinds of different companies. So I think those are kind of signs of decentralization at the infrastructure layer. And as you said, could pair very well with a lot of the stuff that you're working on. And so thank you so much for joining us today, Balgy. Great being here.