Skale Co-Founder Jack O'Holleran Explains Why Blockchain Beats Mongo or Oracle for Enterprise Usage

Audio-Only Podcast

Editor's note: Find the Blockchain Journal Podcast wherever you listen - Subscribe to our content on Apple Podcasts, YouTube Music, Spotify, SoundCloud, and Pandora.                                  

  

           

Full-text transcript of Bob Reselman's Interview with Jack O'Holleran of Skale Labs

Bob Reselman: Hi, I'm Bob Reselman, and I'm here at Outer Edge 2023 at the LA Convention Center in downtown LA. It's one of the premier NFT shows in the country, and I'm having the pleasure of interviewing Jack O'Holleran of Skale Labs. Welcome, Jack.

Jack O'Holleran: Pleasure to be here. Thanks for having me today.

Reselman: Yeah, I've been excited to hear about what you're doing. You're doing a lot in the distributed ledger space, aren't you?

O'Holleran: Yeah, so I've been working on Skale for over five years now, and it's a centralized project where we have over 20 EVM chains that are running in a fully decentralized manner. Our goal is to help people who are building applications for Ethereum, build them and reach a huge audience, not a small audience of developers, but a massive audience of consumers and real users across gaming, and Web3 social apps, and decentralized finance.

Reselman: Great, so one of the problems that's always been around, what we call distributed ledger technology and Ethereum and EVM (the Ethereum Virtual Machine), has been just around the cost of transactions as well as the speed of transactions. How does that fit into your world? Is that something you've encountered?

O'Holleran: Well, that's a problem we've been trying to solve or working on solving, I think doing a pretty good job over the last five years. One, I think, hats off to Ethereum, just the creation of such an amazing technology and pool of users and pool of money and value that's sitting there, and support that's in this network; it's amazing. But, we all know that it's very expensive to transact. And one of the issues is it's both infrastructure and unit economics. And so, there's a lot of different ways people are taking to try to solve this gas issue. We, for example, believe the best route is utilizing a prepayment model.

Applications, businesses, groups of users, organizations, DAOs etcetera, can actually pay in advance for the compute space required to run transactions, which completely transforms the unit economic model. Now, you still can have validators that are running proof-of-stake blockchains, which is happening in Skale, but the way they're paid is different. It's not per transaction. It's like running a car, but you pay for the month instead of paying by the mile.

Reselman: Wow, that's a really interesting way of doing it. For a layman's explanation, how do you manage to pull it off, if you don't mind me asking?

O'Holleran: Well, I'll try to step back first and then we can get more deep, but essentially there are three layers here. The first layer is Ethereum. So there's code and smart contracts for Skale that live on Ethereum that support the Skale network. Then the second layer is the Skale network, which is a whole set of nodes that are run fully decentralized. Now this network actually doesn't even have any blockchains in it. It's a system that's one — that creates chains, that's a third layer. The third layer of the app chains or the Skale chains. Each one of these Skale chains is incredibly fast, has more throughput than the major layer one chains, a lot of them that you read about today and layer two chains. And so this layer three environment or app chain environment, I think, is a future. And there's a lot of people that are looking at this model and saying, "Hey, by decoupling the infrastructure, making different trade-offs for security and performance, this model is one that allows the computation to be, instead of divided and managed in a vertical manner, to be spread out in a horizontal manner, so that there's more even distribution and you can essentially get more throughput and more volume by creating more quantity without losing security." So, happy to kind of dig into that, but at the highest level, those three layers come together to build a unique solution that has speed, security, and decentralization.

Reselman: Right, and it's a different financial model, too, because if we're paying per transaction, that's an operating expense. So, the more business I do, the higher my operating expenses go. But if we turn it into a subscription, or we can call it a subscription model if you don't mind, now it becomes, for the most part, pretty much a capital expense. Is that an accurate assessment?

O'Holleran: Yeah, it's decentralized infrastructure as a service, like our platform as a service. People in enterprise figured, and Web2 figured out things out a long time ago that if you want to charge your users each time they use your product, you're going to motivate them to not use your product. So, instead, they figured out new business models. And [the] same thing, whether it's not [may]be a user, but an enterprise, it was, 'Oh, pay per byte, don't pay per byte, pay per seating, and buy these seats, use an X amount if you go higher then...' And so, you're motivating people to use your product through subscriptions as opposed to demotivating them by a per-event fee.

Reselman: Right, so let me toss you a hardball question here. I'm a fairly good-sized enterprise and I've had success with data storage techniques, excuse me, technologies such as say Oracle, that's pretty canonical or MongoDB or SQL Server, if I go open source, I'm using MariaDB. Why should I change? I mean, I get it, it's working for me, I'm willing to pay the price. Where's my benefit?

O'Holleran: The way I reframe that to people is: okay, well, these databases that you're talking about are one database that's run by one entity that has a guaranteed to deliver accurate information across that database. Blockchains are not efficient. There's not one database. There's lots of databases whose job it is to share the same information and ensure it's accurate, and it's not one party, it's lots of parties. So then the question is, well, that's not as efficient, because there's so many more databases [that] need to operate in harmony at the same time and communicate with each other. It's going to be slower, not as cost-effective because you're having to replicate these databases. So then the question is, why would you want to do that? And it all comes down to use cases. There's certain use cases where you don't want to trust or rely on one single database to store information.

One would be decentralized money, which we all know Bitcoin in popular, and people looking at Ethereum as a store value too, using lots of actors to run servers to guarantee that we don't all have to trust each other, the money's there. But now let's look at business applications. The question is what business applications require decentralized infrastructure compared to centralized infrastructure? And if you can answer that, then you're on the right path and you're not going to be using Oracle or MongoDB.

Reselman: So... do you mind? Let's dig a little deeper because this is fascinating to me. What sort of concrete use cases would you want to use a distributed ledger for?

O'Holleran: Yeah, so I'm going to pretend you don't hear the word NFT, okay? Let's say this is a digital asset that just happens to be an NFT. Let's look at a major sports team. Let's say they wanted to roll out a fan passport that every single fan could have, buy this passport, and there's different levels. And based off these different levels, I could then go and get first access to tickets, first access to season passes, first access to specialized merchandise or meeting the team, etcetera.

Now, you could still do that in a centralized database, but if you're utilizing this in an NFT environment, all of a sudden you automate all, I sell mine to you. And that could happen anywhere, but when that transaction happens, it happens on-chain and that, in an automated manner, payments are sent back to the provider, the team, or the entity on the user side, you have something that's yours. It is on-chain immutable, and it's also a generational thing. The younger generation has really embraced NFTs because they want to own their assets. They also don't want to just, I think just trading collectibles that are scarce, that people are speculating on is not the future of NFTs. That's why I think we need different names for these things. But businesses that want to utilize these principles, activate community, activate engagement, drive for... drive different revenue streams, need to embrace blockchain in order to make that happen.

Reselman: Well that that's good. I like that a lot. Let's talk about the migration path. Okay, so you sold me, all right. I'm CFO. I'm saying, "Okay, you got me. I want... sign me up. I want to use this DLT stuff. " What does the road ahead look like for me?

O'Holleran: Yeah. Well, number one, so I worked in SaaS back in 2000, I think nine. It was when I first started working in SaaS. And half the time I talked to a CIO, they'd say, "I want on-prem if you can deliver me the same solution. I don't want to use Amazon, I don't want to." And fast-forward, even five years later, those same CIOs that were saying they didn't trust the cloud, lost their jobs. And in blockchain, we're seeing a similar path where people are like, "Oh, I want a private blockchain. I don't want these open, decentralized networks." And so now, we're seeing the same transition where the leading edge, if in the enterprise who are adopting Blockchain are recognizing public open networks are the path. One, for their compliance reasons, it's better. But two, they're less brittle, they're more — that's how you capture community engagement.

The values of blockchain are really delivered and open network compared to a closed network. If I have a closed network, may as well just run MongoDB. I don't need to run a blockchain. And so I think the transition path is one, understanding that open networks can be utilized, and then two, building a technical execution plan, and three, having a compliance plan because the legal components of this are still being sorted. And you need to make sure you're operating under a manner that is, I think you have a very good plan. And the thing is, people are lucky now. They're not the first ones to do anything. At two years ago, enterprises were just getting in and they were really having to forage the path on the legal and compliance side.

Reselman: What sort of staffing changes do we need to make?

O'Holleran: Yeah. So one, I think the amount of agencies and consulting companies, and implementation companies that have these core competencies has grown immensely. But that being said, you still want to have your own staff that understands these things. So one thing I've found working with businesses that are launching a blockchain, if you don't have somebody in-house who understands Solidity, the concepts of security around blockchain, the concepts of values of decentralization to the users, you're really at the mercy of your vendors and partners. So it is good to have the internal team, but you now have the luxury of not needing to build an entire team yourself. There are enough really quality teams that are out there ready with experience.

Reselman: Great. And then, of course, there's the other side of the coin: how do I identify a vendor that is other than what they pretend to be?

O'Holleran: Well, I think there's a very high charlatan ratio in Blockchain. There's a lot of hype artists, and I don't want to say scam, but you know, need to look at people's background. And unfortunately, a lot of people with huge depth in of experience and enterprise are not like plentiful. They're not a lot of those people working in blockchain. But there are different, the agencies that are good have experience. So I would not be looking at giving somebody a first shot at blockchain. The security, the issues that can go wrong are too great. So look at the track record most, look at the individuals on the team, get to know them, and look at the projects they have launched. And if they haven't launched, definitely do more shopping.

Reselman: All right. So let's just take this one step further, and now I won't belabor too much, but let's do, I'm a CIO, CFO. I've heard a lot of good things about DLT. We've done the cost analysis, but we understand that we need some help with staffing, and I'm going to interview a vendor. What would be a hardball question to really ask a vendor to make sure they knew what they were doing?

O'Holleran: Yeah, I'd get into... one. I think on the technical side, there are questions around "what are the tradeoffs we're making around security and performance?" The other one would be, "How do you think about user experience?" Because so many blockchain initiatives have been absolutely killed around user experience. I would be asking, "What's your advice on custodial wallets, and should open wallets exist simultaneously? What vendors should we be looking at for those initiatives?" Also, I would be getting at , "What are the values of blockchains providing? "And everyone needs to ask themselves that. There's so many things like, "Oh, let's do it. Let's put on blockchain." But that use case maybe doesn't need to be on blockchain. So I'd be getting into that with those vendors and make sure they understand the business components and they're helping drive that strategy along with the technical side.

Reselman: That makes a lot of sense. I'm trying to think of a hardball question. A hardball question I would ask. And something for me, it would really come down to send in your CTO or your senior engineer, and then, "Tell me about how you're going to take a look at this Solidity code and tell me what my risks are. How much is the solidity code going to cost me to run? And, how do I make it cheaper?" Is that a good question, one might ask?

O'Holleran: Yeah, I think so. And also, "What are your thoughts on audits and how do you evaluate the vendors you work with based off the audits they've done?" Right? Because we at Skale work with a lot of these agencies and providers. We're at the infrastructure level. We partner with some really great teams, and they're responsible to be for a lot of this, the partner side communications, and one of the things we do is make sure we outfit our partners with really good materials on the audits that have happened, the security properties, the performance advantages. Dartmouth just did a big study recently, and Skale was highlighted as one of the fastest and most secure, actually the fastest blockchain in terms of throughput and finality. And it's things like that. They should do their homework. They shouldn't just be like, "Oh, we're friends with this other chain that gave us a big grant." There's a lot of that, unfortunately, in blockchain. There are people that are paying for their business, and I would recommend steering very clear of those.

Reselman: Okay. Great. Well, this has been a great interview, Jack. You really taught me a lot. And I really like engaging in these detailed discussions. So again, here with Jack O'Halloran from Skale Labs, and I'm Bob Reselman from Blockchain Journal, and thanks for sitting through this great interview.