The Pomp Podcast - #390: Ben Golub on Decentralized Computing & Storage

Episode Date: September 23, 2020

Ben Golub is a serial entrepreneur and CEO, who has played a key role in building six start-ups. Three of these were as CEO, including Docker, Gluster, and Plaxo. Today he serves as Executive Chairman and Interim CEO of Storj Labs. In this conversation, we discuss decentralized cloud storage, economic incentives, scaling software companies, acceleration of computing trends, and real world use cases. ============================== Crypto.com is the only all-in-one platform that allows you to BUY / SELL / STORE / EARN / LOAN / INVEST crypto all from one place. Join over 1 million users currently using the Crypto.com app. Download and earn $50 USD using my code ‘pomp2020’, or use the link https://platinum.crypto.com/r/pomp2020 when you sign up for one of their metal cards today. ============================== Coinbase Wallets are adding support for .crypto and .zil domains through their partnership with Unstoppable Domains. Unstoppable Domains provides an all-in-one solution for blockchain domains. You can send money using these new domains instead of long Bitcoin wallet addresses, while also storing your domain in Coinbase's collectibles section. Go to unstoppabledomains.com in the dapp browser to register and manage your domains.  ============================== Pomp writes a daily letter to over 50,000 investors about business, technology, and finance. He breaks down complex topics into easy to understand language, while sharing opinions on various aspects of each industry. You can subscribe at https://www.pompletter.com

What's up, everyone? This is Anthony Pompliano. Most of you know me as Pomp. You're listening to the Pomp Podcast, simply the best podcast out there. Let's kick this thing off. Ben Golub is the serial entrepreneur and CEO who has played a key role in building six startups. Three of these were as CEO, including Docker, Gluster, and Plaxo. Today, he serves as Executive Chairman and Interim CEO of Storage Labs. In this conversation, we discuss decentralized cloud storage, economic incentives, scaling software companies, acceleration of computing trends, and real-world use cases. I really enjoyed talking with Ben, and I hope you enjoy this one as well. Before we get into this episode, I want to quickly talk about our sponsors. First up is

Crypto.com. They're an all-in-one platform that allows you to buy, sell, store, earn, loan, or invest crypto all from one place. You can join over 1 million users currently using the Crypto.com app. All you've got to do is go to crypto.com and download the mobile app where you can buy, sell, store, earn, loan, or invest crypto all from one place. Crypto.com, not only is it a great URL, but that's the place where mass adoption is happening. Go check them out at crypto.com. Next up is Unstoppable Domains. If you've ever sent any crypto, Bitcoin or otherwise, to a long string of letters and numbers as a wallet address, you know it is scary. Your heart starts pumping, you're sweating.

It's not a good feeling. You get really nervous. So now Coinbase wallets are adding support for .crypto and .zil domains through their partnership with Unstoppable Domains. They provide an all-in-one solution for blockchain domains. You can send money using a new domain like pomp.crypto Rather than have to send it to some crazy long random string of letters and numbers, a Bitcoin wallet address, you can now just send it to pomp.crypto as a domain name and then I'll

receive the money. Don't send anything. So if you want to buy one of these domains, you have to go to unstoppabledomains.com in the Dapp browser to register and manage your domains. It's just like any other URL. If you want to buy a normal domain on the internet, you go to GoDaddy and you go and you auction for them, or you bid on them, or you buy them. If somebody else buys it before you, then it's gone. You don't get it. That's it. There's only one. Same thing with Unstoppable Domains in the decentralized world. Once the domain is taken by somebody else, you can't get it. It's gone. No way. It's off the market. So head on over to UnstoppableDomains.com today and go get the domain you want before it's gone. UnstoppableDomains.com. Lastly, don't forget that

I write a daily letter to over 50,000 investors about business technology and finance. I break down complex topics into easy to understand language while sharing my personal opinion on various aspects of each industry. You can subscribe at pompletter.com. Again, pompletter.com. All right, let's get in this episode with Ben. I hope you guys enjoy this one. Anthony Pompliano is a partner at Morgan Creek Digital. All opinions expressed by Pomp or his guests on this podcast are solely their opinions and do not reflect the opinions of Morgan Creek Digital, or Morgan Creek Capital Management. You should not treat any opinion expressed by Pomp as a specific inducement to make a particular investment or follow a particular strategy,

but only as an expression of his opinion. This podcast is for informational purposes only. All right, guys. Bang, bang. I've got a special treat for you today. Ben is here. Thank you so much for doing this, sir. Oh, my pleasure. My pleasure. Let's start with your background. You are probably one of the most educated people who have ever come on the podcast. Let's just talk through kind of where you grew up, various schools you went to,

and then we can get into kind of your entrepreneurial career. Sure, sure. So I grew up in Cupertino, California. So the story I like to tell is in third grade, we had a class visitor who wanted to tell us about this new thing he was working on, and it was Steve Jobs. So that started early,

but my first Silicon Valley job was cutting apricots. So it was like 12 and 13 making dried apricots. And then they paved over the orchard and they built that. So I've kind of been a big of it for a while. But yeah, most of my career has been sort of in startup land. I've done eight startups, four as a CEO. And, you know, sort of started out my career somewhat different. I was actually in demography really in my first year.

I spent time in Africa working on rural health and things like that, then went to Uzbekistan to try and start a business school. Our business case for the business school was that we teach people how to read a balance sheet and democracy and free markets would just happen, and that didn't quite work. We managed to teach a bunch of classes, but in every class, there was somebody sitting in the back wearing a nail-fitting suit, never talking, taking copious notes. And eventually, we ended up getting shut down by the government. So that was my first nail startup. But since that time, I've been kind of fortunate.

I was really in Web 1.0 working for VeriSign with this crazy idea. You can do transactions on the Internet and do it so securely. So I ran the security payments businesses for them, then was CEO of Plaxo, an early social networking company, which eventually we sold to Comcast, then CEO of Gluster, which is an open source storage company, which sold to Red Hat, then CEO of Docker, which sort of launched the container revolution. for those of you who are listeners who are in the computing space. And then most recently, of course, I came on board as executive chair of Storch, which is the decentralized store. So we got to talk about Africa is a continent where most people don't think to go and work. And then Uzbekistan is not the place that most people would think to go build a business.

What was the... No, yes. I said, what was the... genesis for uh for wanting to go to those two places uh well so um uh africa or in particular i was in uh i was in uh kenya and at the time uh kenya had a uh had a fertility rate of eight which means that the average woman living through her childbearing years would have eight kids uh so you know for the fastest growing uh populations on the on the planet um and there were 40 different uh ethnic groups each which was growing at a slightly different rate and

And for political power, it kind of depends on, at some level, on population. So, you know, for me, it was a really fascinating idea. Like, can you somehow take math and caring about people and combine them and find a way to intelligently help people? And actually, you know, Kenya's turned out to be one of the big success stories over the past, you know, past 30 years or so in terms of development. But including, by the way, that they figured out this great decentralized way of doing finance using text messaging. So, you know, a ton of banking and lending and sending money from the city back home all happens via text messaging. So it's a pretty exciting place. Absolutely. And so when you finally came back to Silicon Valley, started working at the technology companies, just growing up there,

I'm assuming that you, one, wanted to work at kind of these technology companies, but two, you had aspirations to one day have your own as well, or was that kind of a, okay. And so let's talk maybe about Docker, which is kind of the last one that you did before moving into kind of the crypto decentralized storage world. Just tell us the story of Docker and kind of how you guys built that company. Sure, sure. I mean, so Docker was a company that actually started life as something else. It was a platform-as-a-service company and used some underlying technology to do that

that was a radically different approach towards how do you sort of encapsulate workloads, if you will. um and um docker was a uh you had this idea that said hey you know the the way that people build ship and run code is it was almost like it was sort of in the pre-industrial age if you will right because code was a long time to write a long time to update it was done by large groups of people code was tightly tied to um the servers that it ran on and docker had this idea of sort if you will kind of

you're creating the shipping container equivalent for code right so that you can take

any piece of code put it in the equivalent of a digital shipping

container and of course the nice thing about shipping containers

in the physical world is that you know it doesn't matter what's on the

inside the outside is always the same right always the same dimension

books are both the same places so in the physical world you want

to ship goods from point A to point B you put them in a shipping

container and it It goes from a ship to a train to a truck to a crane without ever having to be opened. And it can, you know, it's revolutionary. Shipping can revolutionize physical transport and Docker really revolutionized the way that code is written and shipped.

So everything you read now about cloud-native computing and, you know, computing in a massive scale where workloads get split up into lots of pieces and run all over the place, that was Docker. And we, you know, get it kind of in good decentralized fashion, if you will, by building a massive community and being completely open source and, you know, letting a thousand different approaches blossom and see which ones work. And so it was an amazing ride. And, you know, something like, I think the estimates are like 70% of all new code that's written is written sort of using the sort of Docker Kubernetes pattern. what's interesting to me is uh when people think of docker i don't think they think of

decentralization and kind of they don't necessarily see um huge comparison to that and what you're doing with storage but there actually was uh a lot of the same ethos right there's open source there's decentralization it's very community driven um type thing maybe talk a little bit about what you know how intentional was that versus um kind of you got pulled in that direction while building Docker? Because that's just where you guys saw traction. Yeah, no, no. I mean, actually, it was really quite intentional.

I mean, I think we were, if you will, sort of going up against a bunch of really, really large incumbents. Sort of the VMware's and the large cloud providers were all doing things based off of virtual machines and a particular way of doing things. And we knew that a 14-person company is not going to change that right but a massive community can't and so we were really very

intentional in saying hey let's let's build a big community let's let's make the code open let's take a really decentralized approach and we didn't use those terms right but we you know like a really you know distributed decentralized approach was how we build build the code how it gets adopted we built lots of really open interfaces anybody could take the code, modify it, use it in ways that they saw fit. And that enabled us to have a massive community. So it sprang up overnight. I mean, it was one of the largest projects

kind of in the space of six months, if you will, right? And got to the point where billions of downloads a day were happening of code that was sort of encapsulated in Docker. And that is really very similar to the decentralized e-files, right? I mean, that you give up control, that you, you know, you do everything in the open, you make interfaces clean, you sort of do sort of individual empowerment, right? You empower individuals to do what they want to do. You set incentives, right? And then suddenly magic happens.

And that's something we have to do if you're a challenger. Absolutely. And talk maybe a little bit about storage. So, you know, what you did at Docker was very, very disruptive. I think that there's been a lot of people who have said, oh, all these centralized storage and centralized computing companies are the incumbents. Maybe one way to disrupt them is through this decentralized approach. Just why go and kind of help storage and kind of what's the model and how does it work? Yeah, yeah.

Well, I mean, yeah. I mean, you know, so I'll start from the first, right? I mean, you know, death taxes and data are kind of like, you know, the givens in this world. And this past year, the world created enough data to fill a stack of CD-ROMs that would stretch to the orbit of Mars and back. And it's going to grow like 40% next year, right? So it's crazy how much data is created. And, yeah, a lot of that's cat videos, but some of that is really critical information like the cure for COVID or new forms of clean energy and things like that, right? And, you know, for most of computing history,

data was stored locally, so you pushed stuff to a hard drive. And then we had the cloud computing companies starting at the Amazon AWSs and Microsoft Assures and Google GCP. And they actually started out in storage, right? So their first service all of them offered was store your data. You push your data out to there and store it.

and I think they've done a great job but there are a bunch of problems first of all centralized cloud computing is by far the incumbent and something like 90%

of the market is among those three players they also happen to be three of the five largest companies by market cap and there's almost nobody has managed to try and beat them at their game It's such a capital-intensive game, right? So we're sort of in this situation where, globally, more and more data is being created.

More and more data is being stored with three largest players. We've sort of created three Rockefellers, if you will, if data is the new oil. Meanwhile, prices, unsurprisingly, have not come down. In five years, basically, the price of storing a gig of data has only gone down by about 10% in the public cloud. Meanwhile— Why is that? I think it's an oligopoly, right? You've got three big players.

They have no incentive to lower prices because as soon as one of them lowers, when Google entered the market, prices went down and then they stopped, right? Because I think any one of them lowers prices and the others match. So basically, by avoiding lowering the price, you avoid the race to zero.

Exactly, right? So they're avoiding the race to zero. And to be fair, there are, cloud storage costs a lot of money to deliver because you have to build big data centers and you have to power them and provide, you know, fire suppression, all these other sorts of things. But meanwhile, you know, as big as cloud storage is,

you know, it's dwarfed by the amount of storage that's in disks all over the planet. You know, something, again, like 80% of them are less than 25% full. And it takes no more power or people or energy or capital to take all those drives that are out there and, you know, start using the unused 75% capacity that they have. So that's a business idea, right?

And, you know, at some level, we're basically like a massive Airbnb for disk drives only instead of having people rent out spare rooms in their houses, they rent out spare space on their disk drives. So what I want to do real quick is help people understand conceptually how a decentralized storage network works. And in order to do that, maybe we can take two seconds and just describe a data center. And so for those that don't know, there are massive computing facilities all over the world. These facilities basically build the physical structure.

They put computers, they put storage hardware in those facilities. And then they have sales forces that try to go out and get customers. And so that model can be very profitable if done well. But just like any business, when you go to the convenience store, you don't have everyone buying all of the products in the convenience store at all times. So there's things that are sitting idle, or in this world, it is computing power or storage that is unused. And so this idea of a decentralized storage network is basically Ben, the team at storage, has come up with an incentive mechanism to get the unused storage to come up for availability through a digital means. And then people who need storage can go in and basically purchase some of the unused storage at a high level. Accurate. And you're the expert here, but is that direction?

You're really accurate. You know, the only thing I think I would clarify is that the drives that we're using aren't necessarily in in the data center. So we've got we've got some people who have, you know, we've got companies who have data centers, universities have data centers. You've also got lots of individuals who are running computers in their basements. And all of those, you know, people in rural parts of the world that have, you know, have a disk drive, right? All of those can be part of our storage network, and they all are. So we've got something like 10,000 drives all around the world, or 10,000 people all around the world operating drives in like 85 countries. And, you know, collectively, they've created a pretty massive storage network. And if you are storing data with us, instead of us storing it on one drive that we control, we sort of split it up into pieces and we run it across large numbers of drives by lots of different people all over the world.

Not only good economically, it also turns out to be really good for privacy, for cost, for performance, etc. And so the customer experience, from what I understand, is almost identical to what you would do with a centralized service, right? You go to a website, you say, hey, I need to spin up some storage. You basically purchase that. And what happens from that point through a centralized and a decentralized service is different. But for the customer, it's pretty much exactly the same. Yeah, I mean, that was our intent. Our intent was to say that for as far as our customers are concerned, we suggest being really fast, really secure, really cheap, really useful way of doing storage that doesn't require them to change the way that they operate.

So we've got lots of people doing cool decentralized apps on us. We also have massive numbers of people doing traditional apps on us that just happen to be doing it on that decentralized way. Got it. And so then what is the economic benefit? Because that's ultimately the people who have this unused storage, and they're kind of offering it up. I love the example of Airbnb. These are the people who have a home, and they want somebody to come stay in it, right? Exactly.

Talk about kind of the economic difference. Obviously, you're competing with Xero, right? So they get nothing for their unused stuff without you. How much can they make? Or how does that work from you actually sending them somebody to use as a customer? Sure, sure. So basically, if you're a storage node operator, or we call those nodes, you basically download some code onto your drive, if you will. And for the most part, stuff happens in the background.

You don't even have to pay attention. Right. And again, you're not spending additional money on power or equipment or people to do this. What we do as our economic model is we quote prices to our customers. And in essence, about 60 percent of what we make from our customers goes back to them. And so as a storage node operator, we quote you a price based off of how much data you're storing and how much data you're serving up. um and you know depending on the size of your size of your drive and how much you're storing um you know you can make five ten dollars a month for drives right um so it's not you know it's not a standalone business like being a miner

um but for many people it's a great way of monetizing their energy basket um and unlike being a miner it doesn't take a lot of power it doesn't take a lot of effort it It doesn't, you know, we designed this so that it doesn't take specialized equipment to do, that this is an incentive model for, excuse me, for the rest of us, right? And so as a result, we've got a really broad group of storage node operators

who are happily being part of this. And of course, we compensate them in our token, which is STOR change. And so this is something that can be spun up pretty quickly, right? If I have unused storage and I can actually spin it down as well, let's say if I need the storage or something like that also.

It's pretty responsive. That's right, yeah. So now your reputation, if you will, is based off of things like uptime and how long you store. But if you are, let's say, a data center that has equipment that sits idle for most of the year but you need to reclaim it around the holiday shopping days or something like that, that's fine.

We have ways for you to sort of tell us, hey, I need to reclaim my space, and you push the data to us, and you push it to other people. Got it. And so the part of this conversation I'm really, really interested in is I'm going to say, hey, Ben and the rest of the team, they know what they're doing. They're going to execute perfectly, and they're going to end up being really, really successful beyond their wildest dreams. What does this do to the centralized players and how does the market, you know, forces change and kind of the competitive landscape, if that occurs? Sure, sure. Well, I mean, you know, we are able to price profitably to ourselves, right? It's something like a third of what the cloud providers do today, right? And as, and I think as important as the fact that we can price that way over time, economics should help us bring prices down, right?

There's more and more data, more and more drives get shipped all the time, and all the things that we care about, our economics, our performance, our security, they all get better and better the larger and larger our network gets. So what we think, or we certainly hope this does, it actually does disrupt the market. And we'll probably be ignored for some period of time, but eventually I think there needs to be a challenge to the large players. Unlike the other people who try to go into centralized computing, big players like Oracle and IBM and others have tried to enter the computing space, and even they don't have the massive amounts of capital and data centers to compete against the Amazons. The closest competitor that the big three have is Alibaba, who's number six in terms of market cap. right so uh i think the only way that you can compete against large folks and disrupt them is through decentralization and that's true for us in storage but also you know on the compute side as well there are lots of really interesting decentralized companies and decentralized

networking companies yeah and really the advantage you have is you can have a price that is you know 66 percent lower you know give or take because you don't have all of the infrastructure costs and all the expenses that take to build out the storage right you're basically just tapping into unused storage that other people have, and it costs you nothing to actually tap into it. Absolutely, right. Also, we've managed to not only build the equivalent of some really large data centers, but do it all around the world so that we now have the ability to offer cloud storage to people in Kenya.

If you're going to be getting movies served from us, they're going to be coming from Nairobi, not from Nebraska. It's also fast. Absolutely. And then talk a little bit about as you're doing this, what stops them from going into and trying to do a decentralized option, right? Can the Amazons, the Googles, like is that the kind of natural step for them to try to compete as they start to feel threatened by you? Well, I mean, I think it's like most disruptive technologies or the incumbents don't tend to do it because they've got something that works really, really well. And then they have all of their advantages, you know, their data centers and, you know, their, you know, their ability to operate on major thin margins on equipment and their power suppression and their power supplies and their power supply. That doesn't help them in competing with us. Right.

And so eventually maybe they will. But I think at that point, we will have achieved our end. We will have built a great business. And you'll have gotten some incumbency advantages as well. Doing what we do has really powerful economics, but there's also, as you can imagine, different challenges. How do you keep it coordinated? How do you make sure the files don't get lost? How do you make sure that you keep the incentive structure?

How do you make sure there's the right balance between people who want to provide storage and people who want to consume storage? Absolutely. And then talk a little bit about, obviously, as you're doing this decentralized network, people have to get paid, right? And so there's this storage token, right? S-T-O-R-J is what it is. And these tokens have become controversial in the crypto industry. The Bitcoin community thinks, you know, why not use Bitcoin? Everybody else thinks, why not use XYZ token?

Like, just talk a little bit about the use of this payment via a digital token or digital currency and then kind of why create your own versus use one that already existed? Sure, sure. So we, you know, and I just need to sort of clarify, at least the way our economic model works, right now we quote prices in dollars

both for customers and for the snows, for the suppliers. And then customers can either pay us in fiat or in our token. And our suppliers get paid in token and we do a sort of spot conversion based off of the existing price. But there are a lot of reasons

why it's great for us first of all again we're we're now paying thousands of people in 85 plus countries right uh sometimes small amounts sometimes large amounts being able to do it with uh using our token being able to use things like smart contracts really really critical um we've also programmatically built in not only on the supply side but on the demand side so we've uh we built a program we call it the open source partner program that also kind of turns things on the head in terms of demand sides, rather than hiring lots of salespeople, as you talked about the centralized people doing. And then, you know, essentially, you know, what the cloud providers tend to do with the

open source community is take great open source code, give it away for free, you know, as a loss leader to drive more compute and storage networking, we actually compensate the open source companies. So if they send users and data our way, we send money back to them and we do it programmatically using the code. So for us, it makes the system work much faster, much more reliably. Of course, we funded our initial efforts to a large extent based off of having a successful token sale back in May of 2017. And for us, it's the right way to do things.

It also gives us an opportunity to further differentiate ourselves. We're trying to be not only sort of enterprise-grade in terms of our storage, but also enterprise-grade in terms of the governance and management of our companies. Got it. And so do you feel like eventually this is going to force kind of digital currencies, regardless of which one it ends up being, into kind of this entire cloud and storage world where whether you are a decentralized or centralized platform, like they've got to support it? Or do you feel like you've got kind of an advantage and kind of a wedge into a market where that's not going to happen for a long time and you guys will kind of enjoy the first mover advantage there? Yes, it's a good question. I mean, I don't know whether the crypto aspect of what we do ends up being a barrier to entry or not. I suspect that if it does, it's less important than the decentralized aspect of what we do.

We internally say blockchain is much bigger than cryptocurrency, decentralization is much bigger than blockchain, and we think decentralized infrastructure in general is much bigger than any of them. That's really the powerful thing, is that you give tools that push things to the edge, if you will. and push infrastructure into the hands of thousands, hopefully millions of people around the planet. And that's really the big disruption. And cryptocurrency and blockchain, these are all sort of primitives that enable that, but they're not the main story. Yeah, and one of the things that's really interesting just in computing in general to me is the centralized players have been trying to build kind of pseudo-decentralized networks that they control. What I mean by that is all of these POPs and kind of localized data centers and kind of infrastructure that, for those that are unfamiliar with the space, they basically are just pushing infrastructure from rather than having one data center in the U.S. or in one area, how do we kind of break it up and put it closer to the user, but they still control it.

It feels like it's already moving in the direction of decentralization from an infrastructure standpoint, just the ownership has stayed centralized. You guys really are saying, hey, we have decentralized ownership of the network, and we have decentralized infrastructure, and therefore, that's the advantage, right? Right, right. And also we've designed things so that, you know, we're not a central point of failure, right? So, for example, like if you're storing data with us, we don't have any of the keys. We have no way of knowing what's being stored, no way of mining it, no way of compromising it. And the same thing, too, for the storage node operators. So, you know, one sort of clarification about how things work with us, if you upload a file to us to store it with us, it gets encrypted, so scrambled up with keys that only you have. it then gets split up into lots of pieces

generally about 80 pieces of which any 30 can be used for the fact that each of those 80 pieces goes to a different drive run by a different person on a different power supply in a different geographic location and that level of decentralization does magic in terms of security, magic in terms of performance, magic

in terms of economics as well and that's really the key it's not that we have created a network where we push stuff out to the edge we've enabled the network at the end

that's awesome and to me like you know I'm biased because I kind of understood this more and more and more but it just feels like naturally this is where the world has to head right like you guys are just ahead of the curve yeah yeah no I

think it has to right and I think this decentralized approach is really the only way to create a viable alternative to decentralized providers and it's a good job they do And by the way, they offer great services, right? But, you know, again, if data is the new oil, we don't want another Rockefeller. And it's just a bad thing if cloud computing, which is the most important computing trend, is dominated by, you know, three of the five largest companies by market cap. And nobody else can get in.

That's a bad, bad situation. Where do you go from here in terms of, you obviously have the storage network. do you eventually go and compete on other aspects of computing using the same model? Absolutely. Absolutely. So right now, we think that there are lots of great companies that are trying to solve the decentralized compute problem. And so right now, we're in partnership mode. And maybe continuing partnership mode, maybe we can introduce it at some point. There's a long road that we need, a long road map in front of us in terms of taking the storage that we have right now, which is primarily

used for fast backup and things like that and extending it to content delivery and to compliance storage and all these sorts of things. But absolutely, compute, network, and storage are basically the three things that are cloud computing. And if you can provide those primitives well in a decentralized way, that disrupts the market. And we've already found a great way of getting compatibility through providing incentive structure for open source in the same way that we provide incentive structure for storage and authentication.

And so when you built the kind of decentralized storage network, you obviously have to go find these people, you have to explain it to them, you've got to get them onboarded, kind of all that stuff. Would it be fair to say that those same people who are providing storage today could be the starting point of the network for computing or or networking and these other services and therefore it's almost like yeah by building the first network then the second third fourth and fifth get easier and easier over time yeah absolutely right i mean so if you're if you're running a drive uh you that drive already has some compute power in it um and it's got some connectivity so

it's already kind of got like networking and uh computing it now there's there's different technical challenges to do compute securely and to do networking the right way um but yeah absolutely absolutely um uh and uh you know the other sort of there's also a whole bunch of other learnings that you just have to get around how do you run a discipline how do you keep how do you keep supply and demand in in in check i mean that that's one of the i think the biggest challenges that uh you know doing what we do uh has with it in the same way like you know if you're uber or Lyft, you want to have enough cars on the road so that

anybody can always get a ride, but you don't want so many cars on the road that nobody gets a fare. We're the same thing. There's enough capacity out there so that anybody wants to store it in the store, but people aren't sitting idle. Is it also true

on the customer side? If I'm a customer of your decentralized storage, then I'm probably also the most likely to convert on decentralized computing and other services as well. It kind of becomes lower cost of acquisition? I think so. I think so.

What is true though right now is that we have a lot of people who are doing really cool decentralized apps on us and we can talk about that in a bit, but still the bulk of our customers are people who are running more traditional apps. They just want a better alternative than storing the data at the cloud providers.

But, you know, I think what you tend to see in technology in general, right, is that, you know, even the early adopters start with, you know, low, low risk workloads, and when you do a good job with them, then they start going, going upstream. And that's, that's kind of the classic descent, you know, disruptive approach, you, you offer something that's radically better for a small number of use cases that the big folks don't care about. And once people get used to that, then suddenly you get more and more use cases and higher and higher use cases. And before you know it, people who are ignoring you are chasing after you. I was walking you into explaining that you're essentially putting the big guys in the innovator's dilemma, right? Which is... Absolutely, yeah. They can't come do what you're doing because it would be a distraction. It doesn't look interesting, you know, whatever.

but you get to continue to eat the stack of infrastructure on the supply side. And then as you start to source demand, you essentially just eat more and more of their computing needs. And next thing you know, you're back into the centralized players not being in a very good position. That's right. And honestly, they follow the innovator's dilemma too. Twelve years ago, people said, store data in a data center run by somebody else. That's crazy, crazy talk.

These are low-end use cases that move their way up, and we're doing the same thing. If you read Innovative Dilemma, you know that the case study that Clayton Christensen did was on storage because the storage industry is just this history of Innovative Dilemma all the time. Well, I'm cheering for you. I hope that that is where the world is going.

I do too, but it's an exciting time. I think if you're going to go into startups, and again, this is my eighth startup, you've got to do something that you think is going to be fun and meaningful, even if it isn't. Absolutely. Before we get into the rapid fire to wind this down, I want to talk about real-world use cases. And obviously, you have so many customers on each side of the marketplace. Maybe talk through one or two examples of the types of infrastructure providers and the types of companies that are actually using this. Sure, sure, sure.

So I'll talk about customers, people who are actually storing data with us, right? And it's a pretty broad group of people. We've got people who are storing video and photos and things like that, and we give them a fast, easy way to share data. We're seeing a lot of usage among academic data sets right now, where you've got a massive data set of, let's say, climate data or energy data or health data, and you want to be able to share that securely with large numbers of people and make it really fast to get it out there. We're just an excellent solution for doing that because it's very secure, very durable, and yet the data can get extremely fast to everybody who wants to use it. um we're also seeing some more uh you know more relevant to probably the area that you spent spending time and we've got a few people who are actually using us as a way to get blockchain data distributed so blockchains themselves are pretty massive right and

in an ideal world it's really easy for anybody who's running a node uh um to be able to get the entire blockchain and get it quickly um so you know davian classic wrote about how they're using us but lots of others were using us as well um to get uh get data out there and get it uh distributed quickly and securely um uh interestingly part of what you get when you do decentralized storage is you get this massive improvement in privacy and security so as i mentioned you know every file that gets uploaded to us gets encrypted before it gets uploaded then gets split up into 80 pieces each of which goes to a different drive on the network it only knows that they have a piece and so if a hacker wanted to get at that file they'd have to find 80 drives

out of thousands or hundreds of thousands, compromise each one. All they would get would be an encrypted file. And even if they managed to somehow decrypt that, the next file they go after is going to be different. So if you're storing really sensitive personal data, we're a great solution. You remember a few years ago, Equifax got breached.

So millions of people's personal data got compromised because one incompetent administrator at Equifax configured a print server incorrectly. With us, that just can't happen, right? Not only can we screw up, but you can't screw up as an end user, right? You can't do this on stream. So those are the kinds of use cases that we're seeing

is highly secure, highly fast, highly private data that needs to get stored and distributed globally is sort of our sweet spot. And are there situations where companies are choosing kind of centralized and decentralized solutions side by side? So I may be a company, I've got some storage needs, and I may actually split this between both centralized and decentralized. Yeah.

Maybe talk a little bit about that. Yeah, yeah. I mean, so most of our customers right now have some data that they're storing locally in their own data center, and then some data that they're storing, let's say, at an Amazon or Google or Microsoft. And generally speaking, what they will do with us is they'll say, okay, I'm going to give you 10% of my data. and I'm going to try you out. And so, you know, every day as I'm running jobs and I'm creating data, I'm going to push my backups to you or my snapshots to you and see how you do.

And then over time, what they're seeing is, gosh, this is great. Now they're giving you 10% of the new data I create. Let me give you 50, 70% of the data that I create. And then at some point, they may take the data that they're currently storing into Amazon and move it to us, and that will be great. But, you know, what I think is sort of an important truism is that if you're going to be doing something disruptive, it can be disruptive, but you don't want to be, you don't want your customers to experience it in a revolutionary way. You want to give them the opportunity to experience it in an evolutionary way.

Explain that. That is very, very good advice, but explain that. Sure. I mean, we as a brand new company can take on lots and lots of risks, right? But our customers, you know, most customers don't want to have to change what they're doing in order to get advantage, right? I mean, to ask a customer to completely change everything that they do, how they write their code, how they store it, what their risk profile is, is asking a lot, right? And so what we'd like to do is be able to give people the ability to say, hey, I'm going to change three lines of code. Suddenly, I'm using storage, and it's cheaper, faster, better, more secure.

I love it. And I'm going to evolve to do more and more over time with us. And the more they do with us, the more benefit they can get. But asking them to change their world for us is an on-starter. That makes sense. What's the biggest challenges as you guys kind of move forward? What do you spend your time thinking the most about that you guys need to either solve for or kind of come up with solutions? Sure.

I mean, we spent a ton of time making sure that we offered a centralized storage solution that was high availability, high performance, had a large enough number of nodes, ran really well. We actually had a year-long beta process. We didn't lose a single file, which was important to us, but making sure that what we were offering was rock solid. The last thing we wanted to do was be the first to market with a decentralized storage system, have people get excited about it, and then fail. Because we wouldn't just be failing for ourselves, we'd be failing for the whole industry. So we spend a lot of time doing that. We spend a lot of time on incentives. We want to make it really, you have the right incentives in place for people who are running the nodes, and they're profitable and easy, and they're not disappointed by what they have.

And then also that we don't set the wrong incentives, right? We don't want to incent bad behavior. And then a lot of it is, you know, the normal, blocking the chat, building a great company, building a great culture, building a great community. And, you know, that's a challenge, but it's also really fun. Because ultimately, the community is so much larger than the company, and the company is so much larger than you.

That's a great way to look at it. I asked the same two questions to everybody and then you'll get to ask me one to finish up the first one is what is the most important book that you've ever read we talked about one of them already which was the other man in 11 that's great

and lately I I've been rereading man search for me why do you like that one so much I think you know we're living in unusually difficult times them and and i and i think uh you know a you know having the perspective of people who were

facing challenges that just sort of swamp anything we're dealing with yeah it's annoying to have to uh to sit on your butt and watch tiger king but that's not the biggest sacrifice but um you know to understand how to sort of accept the situation you're in and yet still have hope and i think is really really important um and and so you know i think times i guess i find that inspirational and also understand the importance of being kind to each other and having broader broader meaning in your life i think part of how you get through difficult times like we're going through this is to you know work on something that's larger than yourself um so you'll love the second question it's a little bit more fun but man's search for

meaning has a connection here which is aliens uh believer believer or non-believer uh i absolutely believe that there's life teaming in the universe and that there's lots of intelligent life uh i'm not convinced that they've come to visit earth and certainly 2020 i'm sure there's if they are out there they're staying far far away i i've always thought about like do we want to find them or do we want them to find us right usually in at least human history like the the conqueror has has prevailed uh but i don't know in 2020 like this is probably not the planet you want to just jet on over to if uh if you had a choice yeah i guess so yeah the the alien cortez equivalence is looking at us and saying

i don't i don't think so we'll wait till 2021 to go visit 2021 yeah you could ask me one question to finish up what uh what one question do you have for me um so what's next after defy what are you most excited about look i i think um i'm very convinced that centralization has now become a business risk obviously um and so whether that is centralization of your infrastructure like what you guys are going after um centralization of kind of financial services so everything from bitcoin on down to some of the the more nuanced defy stuff um is kind of the where the world is headed. One of the things that is really, really interesting to me as a concept is this idea that we have entire frameworks, currencies, rules, et cetera, for every economy and every kind of

business market, but they're all tied to essentially imaginary lines on a map, right? So they're all like physical location-based type world. And in some weird, potentially wrong, but still kind of exciting world, I kind of think of it as like you and I both left our geographic locations today, and we met in the internet, you know, economy, right? And so we're doing this virtually. And so it's kind of like Ready Player One style, you know, like you're going into this like internet economy or this internet world. And so when you think of it that way, and you kind of take, you know, pre-COVID, all the things you would do, right? So everything from you might go to a happy hour with your colleagues, you had to go to work, you know, you'd go to your kid's ballgame. Like, you kind of just walk through every single, you know, daily activities.

What percent and which ones are going to end up in this, like, internet world? So you're obviously not going to be able to virtually attend your kid's virtual baseball game. Like that might be a little weird, right? At least in the short term. And so like some of it is not going to kind of, you know, be brought into this. But when you start to bring some of this in, so meetings is the easy example for everybody.

There's entirely new businesses. There's all kinds of decentralization, you know, benefits. Like there's just a lot that would change. And I think that for you and I, who, you know, generally are technology minded, have kind of been paying attention to a lot of these trends.

if anything it just accelerated things we kind of already knew were underway but i think a lot about like you know restaurants like all of a sudden a lot of restaurants were like oh this digital thing's gonna happen you know five or ten years from now and like we just got a website right it's like no now now it's here and it got here like overnight accelerated everything and and so you know those are like digital examples i also think that you start to um, break down this idea of like, if companies start to pursue more resilience, um, uh, and not just efficiency, they're going to onshore a lot of things, right? So you're going to see all kinds of manufacturing supply chains, like all that stuff. Um, and so in that trend, uh, American

labor is actually pretty expensive compared to where they've been able to outsource it. And so the unit economics get thrown off. And so I think like there'll be this resurgence of, um, I don't like 3D manufacturing and automation and kind of all these things that allow technology to kind of bring the cost of that stuff back down, but based here in North America. And so when you think of that, like, do we have enough people to do that? Like that are skilled? I don't know, right? And so I don't know. It's just things like that I spend time thinking about. And it's probably intellectually stimulating for me because there's not a right answer, right? So like, I can't be wrong.

So I don't know. But for you guys, right, it's like a perfect, you know, tailwind, right? Because basically everyone's going to need storage. Everyone's going to need computing power. And kind of you'll benefit from this digitization of everything, really. Absolutely. Yeah. You know, I think, you know, I guess other other things, if you're going to go into a

startup, you should, you should be, have the waves moving in your direction, right? I'm not going to go against that, right? I forget who says it but somebody's like uh you know uh good team bad market market wins uh okay team good market market wins right and so it's kind of like you know just make sure you're rolling in the direction of uh of where the waves are going you'll be all right absolutely awesome ben listen thank you so much for taking the time to do this where can people find you on the internet or find out more about storage if they want to go check it out?

Yes, we're storage, s-t-o-r-g-i-o. And you'll go there if you want to become a storage operator, if you want to become a customer, or both. Great place to go.

And I'm at GollumBE on Twitter if you want to follow me. And we've got great blogs and things like that on the storage side as well. Awesome, man. Listen, thank you so much for doing this

and we'll have to do it again in the future. Okay, hey, thanks, man. Thank you.