This Week in Startups - E1098 The Next Unicorns E14: Cockroach Labs CEO Spencer Kimball on the coming paradigm shifts in tech, competing against AWS & more

Episode Date: August 19, 2020

Check out Cockroach Labs: https://cockroachlabs.com Follow Cockroach Labs: https://twitter.com/CockroachDB Follow Jason: https://linktr.ee/calacanis Thanks to our partners... Dell for Entrepreneurs - Save up to 50% off select products, and take an extra 5% off by going to https://launch.co/dell LinkedIn Jobs - Get a $50 credit toward your first job post at https://www.linkedin.com/unicorn Zendesk - 6 months free at https://www.zendesk.com/twist

Season two of the next unicorns is brought to you by Dell for entrepreneurs. It's small business month at Dell. Save up to 50% off select products and take an extra 5% off by going to launch.co slash Dell. LinkedIn jobs. A business is only as strong as its people, and every hire matters.

Get $50 off your first job post at LinkedIn.com slash unicorn. and Zendesk. Listen to Zendesk's new podcast, sit down, start up to hear conversations with Zendesk's leaders and the founders, CEOs, and makers on how to start up, even when the world goes topsy-turvy. Download and subscribe on Spotify, Apple, or wherever you get your podcasts. Hey, everybody. Hey, everybody. Welcome to this week in startups. My name is Jason Callaghanis. I'm an angel investor. entrepreneur writer blogger i guess i used to be called a blogger uh here in the silicon valley boys it's strange it's august it's 2020 we're in i guess i'm in a day 140 or 150 of being sheltered in place i've done a little bit of the distance thing you know went out to get

ice cream with the girls uh on one of these closed main streets here and uh yeah i've been losing my mind i'll be totally honest with you you guys know i'm a social creature i am not designed for this. And I find myself having great sympathy now for people who maybe we're in solitary confinement or in prison. And I realize what a torture it is. And to compare what we're going through in this very mild way to what somebody in a prison goes through, I know is a ridiculous comparison, except for the fact that I very acutely feel lonely and I've never felt sad or depressed in my life. I'm going to be honest with the audience here. I've been kind of feeling depressed about the situation.

And the thing that saves me, and I'm being really honest here with you, the audience, because you've been with me for 11 years, is when I get to do this podcast. Because I get the incredible privilege of talking to the world's smartest, most driven, creative people, the entrepreneurs out there and the investment. who really want to see the world move forward. And I'm very excited about today's guest because he saw an opportunity and he took it. And he fits the model of what creates giant companies. And I get asked a lot after, you know, I've only been an investor for a decade. Before that, I was just a writer, a journalist, and an entrepreneur. But people ask me, like, when a company gets to that kind of like really scale moment,

let's say unicorn status, which is just a number, it's an arbitrary number, a billion dollar valuation. But what it basically means is a very important company, an important company that could change the world. And we have challenged ourselves to say, hey, what are the next unicorns? Now, there's a very easy way to do that. You could just pick companies that are in a certain valuation range, but that's not what we did here. And it's not what we did for the last series. What we're doing here is we're looking at for companies that we think are going to be world changing companies. So the first episode we had Daphne Kolaran. She's from Enceitro.

And if you haven't listened to that episode, it's amazing. She's using big data, you know, and machine learning to try to figure out what drugs we should work on next. Now, if that works, oh my lord, we could figure out and in drug discovery get to solutions in half the speed, maybe 10% of speed eventually. Think about what that would do for humanity. Think what that would do for the edge cases. We could go after diseases that impact a thousand people, 100 people, a million people, as opposed to, you know, the big ones that go, you know, for tens of millions of people. Then we had Nikki Peckiton.

Now, she's from Homebound. You remember that one? That was our second Sunicorn in this next Unicorn series. We're going to do 10 of them. And she's trying to figure out, hey, how do we build more housing? Yeah, that's an important issue, right? We have a housing shortage, and we have. And then we had Gary from Roofstock on, who's trying to figure out ways.

Again, another real estate play. And real estate's a very interesting one because it's hard. Now, what does it take to build one of these giant companies? We're in our fourth episode here. And I gave it some thought. And really, if you're looking for signaling, an obvious one, this is the first one, for a company to really get giant and possibly become a public company that, you know, becomes worth tens of billions of dollars eventually, they need to have raised from the best investors in the world in the private world. That's Sequoia. My boy Ruloff over there.

Mr. Doug Leone, Mr. Michael Moritz, you know, Alfred Lynn. I mean, there's a tremendous group over there. Benchmark, Peter Fenton, my pal Bill Gurley, maybe Bond, a new firm, Mary Meeker, index ventures, you know, like, you know, Danny Reimer. These are the top firms in the world. Well, my guest today has got all of those. So that's like a pretty incredible signal. Like these people have their choice, their pick of the litter, as it were. They need to have worked for some of the best companies. That is a other signal. Now, sometimes you come around at the gates, Bill Gates, it's, you know, Zuckerberg, they drop out of college and they start the thing. But more often than not, they've worked at a Google, maybe they worked at a square, like our current guest today has done. And then

they're in an exploding market. And there is no bigger exploding market than data. At its core, this giant revolution we're seeing, whether it's in biotech, whether it's in social media, whether it's an on-demand, an e-commerce, data is just at the core of everything. That's what our guest is working in. Consumer lock-in, that's kind of important, right? You need to have customers to stick around. Well, databases, people change them in paradigms, essentially. It's a paradigm shifting type technology.

When you lock into a database structure, it's going to be a decade long, two-decade long, maybe three or four-decade-long process, which is why a company like Oracle which has massive competition now, is still crushing it because they have people who've been customers for 30 or 40 years. And then finally,

you need to, and this is a very subtle one, you need to have a passion as a founder for something that other people, a large amount of other people, most people, really don't care about. That you have to be such a nerd

about the thing you're building, that you wake up for those two decades, And on the second decade, you're still passionate about it, right? That's our guest as well. Now, for me, there's a personal one I like, which is clever branding. I love clever branding. I mean, you know, me, buy domain names inside.com, lunch. You know, com.com.

We invested in robin hood.com. We like those great domain names in the world. Uber.com, great domain name. Well, today's guest, Spencer Kimball, is the CEO and co-founder of Cockroach Labs. You heard my long introduction. Welcome to, for the first time this week in startups. Great to have you on the program. how are you doing in this pandemic in quarantine?

How are you surviving? How's your mental health? I gave my overview that I'm losing my mind. How are you doing? Well, you know, thank you, by the way, for that amazing introduction. I think I have to hire you if you ever need a job in investor relations. Yes. Hi, I'm IR for Robin Hood.

Yeah, actually, you know what? I totally understand where you're coming from in terms of the isolation. I think, you know, I felt some of that for sure, especially earlier on. I have also, you know, kind of benefited from the pandemic. And that's a huge privilege. And I don't take it for granted. And that benefit, I think, is because my life, which was previously a headlong rush, has slowed down so that I've been able to really enjoy aspects of it,

like very deep friendships of the people that are kind of. of in my COVID pod, where previously I'd see them, you know, infrequently because 30% of my time honestly was traveling every week. And that's a pretty... It's so weird to recapture all that, right? Yeah. It's like taking things back to maybe an earlier time in my life or a simpler time in sort of civilizational history. And that's allowed a lot of self-reflection, which I hadn't had the time for. And I think I've come away a more thoughtful person than when the pandemic started. How are you thinking about haircuts? It looked great, by the way. The beard is very strong.

The haircut is right now. Let's talk about the beauty. Have you gotten a haircut in quarantine? You look like you trim the size. You know, I have gotten. I've gotten two haircuts actually because Whoa, whoa. Take it easy there, Spencer. Outdoors, I hope. Or is out of doors. No, it was outdoors. although a lot of people at Cockroach Labs did self-administered ones, which were pretty funny. But I think these days you can get a haircut in New York City at the actual salon now. You know, it's obviously people are masked up. As somebody who was born and raised and bred in Bay Ridge, Brooklyn, lived in West, West, West, West Chelsea on the West Side Highway in 26, Tribeca.

That's my, that was my New York hoods. Where about in New York are you? I actually live in NoHo for the last 15 years. NoHo is what? That's north of Soho. North of Houston, yeah. Soho, south of Houston. And it's right between Greenwich Village and the East Village.

Got it. Yeah, I know where you are. I was trying to think there used to be an Algerian restaurant there that we used to go to at 4 in the morning. That was incredible when I was a kid. But I doubt it's still there. What's the vibe like in New York, by the way? I think it's improved dramatically. I mean, there was some dark days in New York.

You know, in March and April, whoof, it was incredible. You could go right down Broadway in the middle of the day and there wouldn't be a car on which is pretty unusual for New York City. And nowadays, though, especially with the summer and the case count remaining relatively low, this good energy in New York. There's not a lot of tourists, so it's all locals. That's always nice. Yeah, the restaurants have all spilled out onto the streets.

They're building all these little seating areas and things. Yeah, that's, it's actually. The way it should be. Permanent. Shut down St. Mark's Place. Yeah, there's a lot of talk about that. You know, why shouldn't this be the way New York is? Why do we have to have so many automobiles in the city?

Makes no sense. It makes no sense. I mean, have you been in Santa Monica Promenade? Yes. Like the Santa Monica Promenade, I looked it up. And there are historical pictures of it with angle in parking. And there was a big debate. Like all those vendors were going to lose business because the car, you know, the two

parking spots in front of them were going to be gone. And it's like, no, it's going to become a destination. Yeah. Well, I'm glad everybody's safe there. My parents are there. My brothers are there. I got a lot, all my family's there. And it's just nice to see the turnaround.

And it says there's definitely something going on. Listen, I'm not an expert here. But we've definitely turned a corner. They're doing massive testing there, huh? Yeah, a lot of testing. And, you know, people I think are pretty good about the mask wearing, which obviously helps. And they, I think, made a fateful decision not to go back to indoor dining. So smart.

Yeah, it's kept the case count low. Because outdoor, I think. Subways, too? Subways, people taking them or not? Because I would knock it on the subway. I rolled out my bicycle, which is usually in storage between Burning Man every year. And it's like an electric bike. I have like a little seat for my daughter on it.

I got the electric. Oh, you brought your kid. Did you bring your kid to a Burning Man yet? No, not yet. She's only three, so one day. Okay. My 10-year-old wants to go now. And there's a group of, you know, folks who bring their kids for the first two or three days,

you know, before it gets too crowded and then, you know, they pop out. All right, listen, the way I wanted to structure the interview today, because you've done so much in your career is. I wanted to start out in part one and talk about databases and how they've grown over the years. And even touch maybe a little bit on open source and the impact that has had vis-a-vis versus Oracle and then building a company off of open source. And I want to get into starting and running a company. And then as we wrap up, talk about funding, which you've done just a tremendous job on. There's a couple tiny things I want to talk about around privacy and big data.

So that's a structure for everybody listening here of how we're going to run the podcast. If you're using an advanced podcast player like apples or overcast and you swipe left on overcast, you can see the chapters. So if you wanted to jump around, you could do that. And we're going to get through a big long list here with one of the smart folks in the industry who's worked out, like I said, Google and Square, and raised from Sequoia, Bond. Just everybody. All right.

Stick with us. We'll be right back. All right. It is small business month at Dell, and that means you can save up to 50% off all Dell products. Here's my Dell Chromebook with all the plugs. I love. I got my USBC. I got my SD card. I got a SIM card. I got an HDMI cable and I even got an Ethernet cable. My studios geared up with Dell equipment because Dell equipment is amazing. If you go to launch.co.com slash Dell, you're going to get an extra 5% off and more information

about the Dell for Entrepreneurs program. And they're introducing their new 15-inch latitude 95-10. A lot of you've been talking about this. It's the world's smallest, lightest, and most intelligent 15-inch business PC. And this new latitude includes the Dell Optimizer, which is a built-in AI that's going to help you with your audio settings and your battery optimization, and it adjust to the way you work, wherever you are. And it's also their longest-lasting PC with up to, I mean, this is getting ridiculous. 34 hours of battery life. The Dell for Entrepreneurs Program is really unique in the entire industry, and they are going to help you level up all of your tech hardware. It was created to support founders.

by providing resources and tools. Those include free IT consulting, because you probably don't have an IT person on staff. It's probably you, the CEO, or the founder. And you get rewards like 6% cash back on all Dell products. They've also got Dell finance and a bunch of other stuff there. It is Small Business Month at Dell, and that means you can save up to 50% off all Dell products.

Check out launch.com.co. For an extra 5% off right now. That's the best deal you can get. Launch.com slash Dell for an extra 5% off and more information about the Dell for Entrepreneurs program. Thank you to Dell for making great products. I have been buying them for decades. Let's get back to this amazing program.

All right, Spencer Kimball is here. He is the CEO and co-founder of Cockroach Labs. Now, you've been working in databases for a bit. I know you were at Google during the formative years when I was running AdWords on my blogs and gadget and auto blog. We were the first case study. and you worked on the database team, I believe, on the AdSense or the AdWords product side? AdWords, yeah, actually.

When it was just getting started and really having incredible growth, I was tasked, you know, partly with, you know, a group of other people with solving some of the problems they were having with their early choice for databases. And what was the early choice there? Were they using MySQL or something? They were. They were using MySQL. And they'd started off, of course, with just one instance. And then as typically happens when you hit huge scaling challenges, you do what's called sharding. At least that's what Google called it.

And I think it's a pretty common term in the industry. But what that means is that you might start with, you know, say 10,000 customers. They all fit in your first instance of the database. And then you say, well, we're kind of right at capacity. We need to add another 10,000 customers because it's growing so fast. So we'll get two databases. And we'll start to put the, you know, 10,000 1, 10,000, so on up to 20,000 on that next database. and then you get to the next one, you add another shard and so forth.

But the problem is that that sounds fine and it actually is a pretty good linear way to scale up, but it creates complexity. So the application doesn't just talk to one database anymore. Now it's talking to, let's call it, you know, two or four, or eight or 16 or 32. And I went through all of those sort of... Iterations? Oh, my Lord. And each time there was like an additional complexity.

Like there were things that just would really break and you had to fundamentally solved the problem by coming up with a new piece of technology that you can fit in there that would help ameliorate the situation. And I think it's important for the audience, especially founders to understand the evolution of technology. Before that, you weren't, you didn't need to have strong response time from a database because in a commercial database, people would go to the finance office and say, hey, I need this report.

And they'd say, when you need it, buy this Friday. And then we'd run a report. And if it took an hour, if it took five minutes, it didn't matter. You just got the report together. You handed it to somebody. But when you're dealing with something like customers who are buying ads, they need to have a dashboard. And back in those days, the dashboard would take a minute to load or whatever, 30 seconds to load. But this is a distinct challenge in a real-time environment when you're displaying ads, when you're trying to do dynamic things.

And when you've sharded, shout out along came polly. When you've sharded your database, it's sharded. But you guys can look up with the reference if you don't know what it is. Great. You've seen Along Came Polly? Rain Dance Rest in Peace Philip Seymour Hoffmourne

who used to work out at my gym Chelsea Pierce by the way and he literally looked like the guy from Along Came Polly because he was a large guy

who wore clothes that were too tight and he'd just be on the elliptical sweating like he he was not playing a part he was literally playing the guy from that film but just you know in my I was never a database

manager did a little IT in back in the day but am I correct in that the issue becomes when you shard databases, now you've got multiple copies of data. Now, if I want to query that data, that data may exist across multiple databases. It has to be found as well. And back in that day, the developers who are writing the code, now they used to point to one database. Now they're pointed to two, but oh, wait, did we update that edge case code for that special thing? And now it's pointing at 32.

that problem was basically you couldn't even throw hardware at it, which is what people tried to do, right? They tried to throw hardware at it. They tried to throw high speed fiber connections at it. They tried to throw memory at it, caching. None of that worked because it just got too big, too quick, correct? Yeah, I think that's a pretty good summary of it. And there's all of those things you mentioned, they help. But when your thing is growing exponentially, you can't just help.

You have to get back to fundamentals and redesign and really plan for that kind of scale. What was that fundamental change that happened in databases around that time? Because we also remember dynamic sites, friend feed, this Twitter thing, other things, just constantly having their databases crash the servers. Obviously, MySpace was a complete disaster. The reason it died was really because of their database structures and choices. what was the big breakthrough from MySQL to non-structured databases, etc.? Yeah, that's a really great question.

And to kind of to your earlier point about how much friction there is once you've already adopted a database, AdWords stayed on that MySQL path. They were at 32 shards or so by the time I left a project. And every morning there was a war room for like months where we were putting out the fires and things. But it kind of got a bit stable there. I moved on to some other things. But they didn't retire that system for 10 years.

And by then, by around 2012, 2013, 14, it was already a thousand shards. Oh, my Lord. Probably the total amount of engineering time was probably, it was probably measurable in the centuries that went into just wrangling that database from one shard to a thousand shards. But that became an anti-pattern of Google. So, like, obviously, you start on a database, you're stuck on it. So Edwards was on that thing.

There were, like, points along that. journey where they tried to move it to other emerging database technologies. And it was always, nope, this isn't ready yet. It's not ready. We're so invested in the MySQL schemas and the requirements that a relational database, you know, the guarantees relational database gives. But it became an anti-pattern of Google. It was fairly outlawed, I believe, just because of the immense struggle that engineering

had to do to make that paradigm work. And they instead decided, and this was kind of like some of the, you know, brightest minds at Google, Jeff Dean and Sanjay. Gamowat. They created what was called Bigtable. Bigtable launched the NoSQL movement. So things like MongoDB and Cassandra are, and H-Base are examples of things that were inspired by Google's efforts. And they had a paper in 2004 about Big Table. And Big Table was amazing. What it did is it said, okay, we're going to redesign databases. So these things scale automatically. And they scale horizontally. So it's not just about getting bigger machines, which, you know, you can only

go up so far before you're kind of at the supercomputer stage and can't really get any bigger and your cost curve is super linear. So you're paying more every time you're just adding it, you know, the same amount of additional capacity. So it's got a low ceiling and a super linear cost structure. That's the old way of doing things. That's sort of the Oracle way. Yeah.

Big table was like, okay, we're going to use commodity hardware in these data centers. You put another machine in. You get that much additional capacity. And the system itself will manage the growth. It will rebalance things, a little reach equilibrium between all of the nodes you have. But in order to do this, they said, hey, you know, SQL is a, decades-old technology. It's extremely evolved. It's got a huge footprint. We're not going to try to build SQL. We're actually going to say, they didn't call it no SQL originally,

but they're going to say, okay, we're going to make this a simpler API. It doesn't do as much. It's not going to have transactions. But what it will do is it will handle the immense challenges we have around scale. For everything Google needed to build, we have, they had a huge scale problem. So that was the number one thing. Can we make this thing elasticly scalable? And they did. And when they first started talking about that, this becomes a paradox. shift because you can't do the SQL queries that everybody's used to, right, forming SQL queries. There's some whole business division. There are some, you know, data jockeys who spend their entire lives, you know, making SQL queries and figuring things out. And they're like, wait a second,

that's going away. My entire career is based on that. How does a paradigm shift, like NoSQL, you know, inside of an organization, even a progressive one like that? What's the religious war that breaks out between the paradigms because I remember when I was a psychology student, they told me, listen, you know, Freud and Skinner and behaviorism and cognitive and young, these paradigms, they only died when the people who believed in them die. So paradigms don't die, people do. Literally, paradigms don't die people do. So obviously the people who were in my SQL camp didn't die at Google, but this religious

war broke out, correct? So Google was a pretty interesting anomaly in the sense that the four years before I got there, they were rarely using any kind of established database technology. They built everything from scratch because they were trying to index the entire web. And that was their original use case. But they started adding use cases. And AdWords was an obvious one where you actually had financial ledgers, right? You were dealing with all these very discrete bits of normalized data.

And relational database made sense. So the people that got on that project were the first people to use a relational database, SQL at at Google and they had scaling problems and that kind of led to a let's not do this again, right? But what's interesting is that the use cases that started to expand from Google's search dominance were not ones that really cried out for relational databases. So there wasn't a culture of using relational databases that was going to be at loggerheads with this new idea of no SQL. Instead it was like, we need scale is the number one thing and this this answers the problem. But interestingly, as soon as big table came out, everyone was like, okay, let's use this for AdWords.

And the AdWords people, and this is very interesting, be like AdWords people, they said, okay, that sounds great. And they started looking at it and they're like, wait, this doesn't have transactions. It doesn't have schema management. We have like a, we already have like 500 tables that are all, you know, we're doing these complex queries between them. This isn't going to work. You know, you guys have to be, you're smoking something, right? This isn't going to work. So it was kind of like, interestingly, back to the drawing board pretty quickly for the big table team. And they created something called Megastore two years later. When we get back from this quick break, I want to wrap up that like transitional process and then talk about building a company around open source,

because this is another, I think, big paradigm shift. And the venture capital community, let's face it, when they looked at this, they said, wait a second, you're not going to hire a bunch of sales executives to take people to the Knicks games and to the Warriors. players games and out for stake and golf, you're going to try to sell to some nerd database people and then build it up in the organization. Are you effing crazy? I want to know about that paradigm shift and then how you raised all the money and your philosophy of building an at-scale company when we get back on this week and startup. All right, I want to give you $50 right now off from LinkedIn Jobs, which is the greatest place for you to find talent. I have found so

many members of my team through LinkedIn jobs. And I'm going to give you $50 with a special code at the end of this quick ad read. And I just want to tell you about a testimonial from one of you, one of my listeners. Aaron Mason emailed me and he said, hey, JCal, I used your code. I got the $50 off. And he is the founder and CEO of Emma AI. That's a startup that uses AI to optimize travel time for people's work schedule. Makes total sense. Great idea. Well, he recently hired a machine learning engineer who started this July in 2020, and this person has hit the ground running and has changed everything for the company. But here is the actual brass tax. A hundred and ten relevant applications came in in four days. Now, let's pause for a second. That's over 25

qualified applications a day for Aaron and his company, MAI, from LinkedIn jobs. Because of the $50 I gave them, you are going to be able to hire great people. people because you all know LinkedIn has over 690 million members worldwide. And who knows, you might be working from home a lot. You might be open to everybody from around the world. You're going to go on LinkedIn and you're going to screen candidates who have the hard and soft skills that you need. So when your business is ready to make that next grade higher, when you want to get that machine learning engineer, or even an SDR, or maybe your director of growth, all of those people are sitting there waiting for you on LinkedIn. And if you go to LinkedIn.com,

slash unicorn, I will give you that fitty, that $50 to get hundreds of qualified people. It's going to be great. You're going to fill that position and you're going to be back to work. Okay, let's get back to this amazing episode, LinkedIn.com slash unicorn. All right, welcome back. It's our second time we're doing this, the next unicorns. And we've really challenged ourselves to find companies that you, our general audience, may or may not have heard of, probably haven't heard of a lot of these companies. But here in the venture community, when you start seeing two or three really notable investors piling on, these are the companies that people start to pay attention to, and they start to get on an IPO path or maybe an acquisition offer start coming in. And today we have

the fourth in the series. We do 10 of these at a time in the next unicorn series, thanks to our partners who support it. It gives us the ability to do this research and find these great companies to share them with you and the lessons they've learned. Our guest today is Spencer Kimball, and he runs a company called Cockroach Labs. Their Twitter handle is Cockroach, or Cockroach, DBA. And what's the URL? I don't have it here written down. Cockroachlabs.com.

Cockroachlabs.com. There you go. Bit of a crazy branding. We'll get into that in a moment. But when these new database paradigms come out, everybody starts moving to them, it becomes clear that startups, don't like to pay for software because they're under-resourced. So you get all these tiny startups engaging in this technology, correct?

That was the thing that happened in the, what, the 2003 to 2010 time period in databases? Not just the startups, but the entire audience of developers out there. It's a 10 million or so of them. They certainly don't like paying for software, even less so than the startups. Which is pretty crazy when you think about it. I'm not sure. When did you start in technology? When was your first job in tech?

year-wise? 90s? 1993. Perfect. Same as me. I graduated Fordham University in 1993 after five years at school at night. And in 93, you remember, you know, if you went into a company, the way software was sold was to the CTO or the CIO.

And then they would come down from on high and say, we are locked into Microsoft, Oracle, whatever it happened to be. And here you go. You're going to training. And that was it. Developers had no power. hour. But all these rabble-rousing open source folks, they started to apply it to startups, they started to play with it on the back end, but that still required buy-in from big companies,

which didn't come till later. But a company like Google, I think, was pretty big on this. How do big companies like Google look at the open-source nature of this in terms of the competitive advantage? Because we used to look in technology as the software as the advantage. So by open-sour, sourcing it, you're kind of giving away an advantage. How does Google decide, or let's just leave them out of there since you work there, I don't want to break any NDAs, but how do the big companies decide when to deploy and invest in open source and when to keep it under the hood? And how do they, how do they maintain that balance of, hey, we found a serious advantage here for our business versus, hey, let's give this to our competitors? So you're talking about big companies that are actually

open sourcing their own software, as opposed to using open source software out there. I mean, it's pretty obviously why they use it, right? It's just a much faster way to adopt new technology. It used to be, as you say, like top-down selling. You'd have to go through procurement if you wanted to use something. You'd get printed manuals. There's no community online. So open source just made a much easier form of consumption, consumption for developers. And when you're talking about a big company open sourcing their own stuff, I think the realization, and it's been proven out time and time again is that the advantage you get from software is difficult to maintain in this day and age because if you're not open source, someone else is going to come along, fast follow you,

and create something open source that's going to, you know, sort of knock the bottom out of your sort of profit structure that you have around your closed source product offering, whatever it is. And you see this happen essentially in all areas of tech. So there's not as a defensible position in being closed sources, there used to be back in the sort of Microsoft heyday in the late 90s. And so it started to get eroded fundamentally in the aughts. And, you know, in the database space,

that was MySQL, right? Yeah. It was a very interesting moment. I just remember these incredible religious, you know, debates and in the aughts where like there was really a lot of fud, fear, uncertainty, and doubts, you know, by the incumbents who were like, don't trust your business to this. And then it was like, but that's the fastest growing business in the world.

and they do. So they kind of, but in summary, the benefit of keeping it close to the vest is not, is, is so small when compared to the benefit of it being out there publicly that inevitably, if you try to do it privately, somebody is going to realize, hey, this needs to be open-sourced and you're just, it's going to happen anyway. So you might as well, cross the chasm yourself and just deal with, you know, giving up that little minor advantage for the, for the bigger protection. I think the huge learning from,

at least my whole career in the past 20 years, has just been that if you have a platform that has committed users, you have the eyeballs. It doesn't matter if it's something like Facebook or something like MySQL or Redis or, you know, cockroach for that matter. If you get that audience that's committed to your product, you will be able to monetize it one way or the other.

And you're talking about an audience that could be orders of magnitude larger with open source than with the closed source software because it's so easy to adopt. And there's just so little friction from the perspective of the developer. Developers are cheap and they're lazy and they want to get things done as fast as possible. And open source really plays to those sort of character traits of the average developer. Yeah. And frankly, I think that the reality is that open source software is easier to trust because you have so much more control over your destiny. You're not paying for as much for it

in the first place. You can look into the code. You can make your own changes. And that that fosters a sense of community, trust, sort of control of your own destiny that is also very appealing to the average developer mindset. So it just, you know, I think it's, it's definitely eaten the software world. It almost is in a way like an advanced form of governance where when the Greeks, when we created democracy and gave it to you all the rest of the world. I'm speaking to everybody to listen to the podcast. We gave you democracy, a great gift. You got to take democracy and you got to play with it a little bit. You could have, you know, regional, you know, United States kind of set up. You have a parliament. You could have executive branches. You can fork it if you like. You can edit it if you

like and that does build trust. And what is the state of the art in terms of governance when you are running something like Cockroach DB. And if I was some, I don't know, bad actor or just, let's just say, cutthroat adventster, and I said, I'm going to throw a billion dollars and I'm going to take over this open source project. I'm going to put 100 developers on it. And you start thinking about, like, in the Bitcoin world, there was always this 51% attack kind of, you know, oh, if you had 51% of the voting, you could start voting it and you could do all kinds of stuff that would maybe we crashed the system. How does one, when you have an enterprise like yours that is destined to become a unicorn many times over in my mind, how does one mitigate against this concept of another

party coming in and trying to control the governance or just taking the entire thing and forking and saying, I'm putting 500 developers on it? We actually had a Chinese company forked cockroach back in around 2016. It's called PINCAP and they have a competing database called TIEB. And we actually, we like those guys. I think they have a great product. It's a bit different from cockroach in terms of some of the, you know, differentiators that it has. But, you know, I think that's what you expect from cockroach and no invitation is the sincerest form of flatter. I do think that's true. Ultimately, in general, with an open source project, especially a complex one like cockroach, having the core committers kind of means everything. And it's very different.

Define what a core committer is for the audience that it doesn't understand this in this interesting open source world. These are typically the people that are either maintaining it or even original authors that continue to work on it that just understand the depth and complexity of the system at a level that, you know, has taken them years. And so to just throw money at something, it's kind of, you know, how many cooks can you put in the kitchen sort of problem? There's only so many people that you can use to effectively take over a project all at once. You can throw a billion dollars out and you can waste that billion dollars pretty quickly. And you might not get anything better than, you know, a steady state, you know, cost of spending like, say, $20 million.

of the same time frame on the core commitors who already understand the project. So that's been the typical defense of someone just coming along and truly building another business. But there's been some interesting developments, which I can get into if you're Yeah, well, definitely, we definitely need to talk about China. But a way to think about this would be, you know, if you were to copy Star Trek or you were to copy a genre like, you know, Star Wars, you don't get Gene Roddenberry and you don't get George Lucas and the creative team around them. So there's been a million Star Wars or Star Trek knockoffs, you know, for better or worse. I mean, The Sopranos comes to mind as like a really great homage to Scorsese and, you know,

the godfather, right? So you can build on things successfully and you can have, you know, those actors and folks move on from them. But it very rarely results in something transcendent that unseats. The Sopranos only reinforces the brilliance of the Godfather. And, you know, the Goodfellas only reinforces the Godfather's brilliance. When we get back, I want to talk a little bit about China and data and specifically privacy. And we'll get in the final act into what you've done in terms of going from a CTO of a aqua-hired company to now the CEO. And how you made that transition of one of our Sunicorns. here on this week's startups.

All right, listen, everybody knows Zendesk is the go-to tool for customer support. You know that. It's the standard. It's the gold standard. But what you may not know is that Zendesk also offers a suite of sales tools, because that's part of the process, right? You sell into a customer. You delight a customer.

Then they have a challenge. You want to make sure you solve that challenge, get that great information, and then put it back into the product. This is called the flywheel. You know, you got sales. you got product development, you got customer support, and all of that is done through Zendesk. They have a beautiful suite of tools. Zendesk is offering their suite of sales tools, plus their industry-leading support software, which you need to be using for free for six months,

as part of the Zendes for startups program. And I'm going to give you the secret URL to do this in a moment. You're also going to get Zendesk's community of startup founders and partners to help you better serve your customers, right? You're going to learn from your peers. And they'll even offer dedicated onboarding guidance and support to get you up and running. And your company is going to be stronger. Like my company, one of the companies we invested in Steasy Studios, which is a dance app through the combination of Zendesk, Explore, and their ticketing, tagging system, Seasy can track, which features their users are most excited about. And they give that to the product team. Bing, bang, boom, all of a sudden, the product gets better.

six months of Zendes for startups for free at Zendesk.com slash twist. Okay, you may have guessed a URL. Zendesk.com slash twist. If you're under 50 people, if you're Series A or under, they're going to go ahead and give it to you for free for six months. And check out Zendesk's new podcast. Sit down, startup, available on all major podcasting platforms. Every customer counts when you're a startup, especially now.

So start building the best customer experiences with Zendesk. Okay, let's get back to this amazing podcast. Okay, Spence. Kimball is with us. We're on a first name basis right now. You guys call him Mr. Kimball or Spencer. I can call him Spence because, I don't know, I spent 20, two minutes with him on a podcast. And we're both losing our minds. You grew up in New York, by the way? Where'd you grow up? I grew up in Jersey mostly. Really? We're at Ridgewood? Where are you? North, south? This north-central kind of Maplewood is what it's called. Oh, okay. Very nice. Very nice. And then where'd you go to college? I'm just curious. You see Berkeley. You went to Berkeley. From Jersey to Berkeley. Oh, my Lord.

I wanted to go to California, no matter what, I thought it was going to be the promised land that would look like Baywatch and everyone would be playing beach volleyball. And I went to Berkeley. I'd never been there before. Slightly different. Slightly different. Yeah, they're not, they're not taking up, yeah, volleyball. They're taking shrooms. It's a slightly different vibe. It kind of results in the same transcendent experience, paradoxically, but it's slightly different.

All your family and friends gave you shit from going from Jersey to California. You start calling California boy asking you if you got a convertible? No, I don't think, I think they were mostly jealous. They're a little jelly, yeah. It's something else. When you move from that northeast to California, everybody's just like, oh, you defected. But then they're like, what's it like? Is it actually great?

And it's like, pretty great. When I moved back, they were like, what would you move back for? Why did you move back? What was you thinking there? You just missed the energy of the city? You missed that vibrates of New York? I'd only worked in summer jobs in the city previously. And so I'd never lived here, but I came, Google opened an office

for engineering after Bell Labs started basically hemorrhaging these really qualified PhD candidates. So I came out to help and I stayed for three months and fell in love with it. And mostly it's because New York had such a vibrant social scene compared to San Francisco. Would you move out there in the aughts or so?

Yeah, 2004, I moved permanently. Oh, yeah, yeah. Post 9-11, it was the city was in transition during that time. I left in 2003, 2002-2003, but man, you missed the 90s in New York. It was crazy. That was a good time.

Anyway, I caught a little of it. You know, I saw the meatpacking district. And when I first came here in 2003, it was, you know, blood on the streets. Literally, I tried to explain this to people. There were literally two places you could go in the meatpacking. There was a club coal Mars by peerigation. Yeah. And it was in a five or six-story warehouse building.

And be careful on floors three, four, five. Like, it just got more and more deranged as you got to the top. And this was like a lawless time. New York. Second, there was a place called Florent, which you might remember, which was a French bistro that was open 24 hours a day.

But that was a time when there were literally, let's just say people working on the streets in recreational practices in the meatpacking districts. And there were two places open that somewhere around 4 o'clock, 5 o'clock, we'd be going to get stakes at Floreon.

You go get the steak fritz there, steak au au la Poiv was on point. And the meatpacking would open. And you would just see guys in butcher coats with cows on hooks running down the street. It was a literal meatpacking district in the 90s. And if nobody here is, if anyone that's listening hasn't been to Manhattan recently or to the meatpacking, it is a very different scene nowadays. It's literally like Times Square.

I mean, you have the best hotels, the Apple store. It's unbelievable. Every nook in crannia, that city has been. Yeah, it's like, it's gilded right now. It's posh beyond belief. Well, I think it's, I don't know how you feel about it, but I think one of the upsides, I always look for the silver line and try to be positive and creative destruction, the creative destruction being caused by the flight of families and professionals who can now work remote out of New York City and San Francisco is going to result in a collapse of rent, which then results in people who have commercial space letting people like me in the 90s. I lived in a commercial space for like $1,700 bucks a month, two or three thousand square feet looking out on the Hudson River, a giant long. And I just built my own bathroom and it was completely illegal. And that was like no big deal. You just paid the landlord $500 a year at Christmas. You were good. And that's what's going to happen.

The artists are going to, I heard people, Brooklyn's more expensive than Manhattan. That's true, I think, in a lot of cases. And they're going to have to redeploy all these elite spaces and try to figure out how to get people back into the city. The rents are going to collapse. San Francisco right now? It is a disaster. I mean, I lived in New York in the late 70s and 80s. Whoa, I mean, it was dicey.

Let's talk about something dicey in your business as we get back to cockroach. I could talk about this city thing forever. But, you know, you have an interesting cold war occurring between the United States and China. Most people would believe a communist country. Having access to a lot of data is a liability, putting aside how you feel about our current. management here in the United States, I don't feel comfortable with social networks or networking hardware, Huawei, or TikTok being anywhere in my house or office for obvious reasons. In your mind, looking at the situation with TikTok and without having knowledge, any social

network, let's put TikTok out of the situation, any social network that is based out of an an authoritarian country, the access to the data in that database would be de minimis, trivial to have access to, correct? If you had access to the employees and, you know. Yeah, listen, I would, I would, if I were an activist from formerly living in Hong Kong, I certainly wouldn't want any of my personal communications being stored anywhere near Chinese, you know, access or, you know, any of their networking equipment. Yeah. And so the idea that we would allow a piece of software on 30, 40% of phones with access to location data, the camera role, location, photos, your microphone, your camera, your address book, this would be, if you were to say, I'll give all of that information on, I don't know, 20% of the population to an adversary that has a leader for life putting a side of its China.

anybody, that would be a crazy prospect, no? Yeah, I think so. I mean, the reality is there's, there's just no way to protect this data from these, you know, for long, I should say, from these determined state actors. And, you know, I suspect that the people that run TikTok are probably, you know, pretty honest and well-intentioned. And it doesn't even require them to build in back doors. I mean, the reality is there's these zero-day exploits and things that a state actor

like the Chinese government can get any. into your phone if they really want to. Explain zero day for folks who don't know. It just means that there's nobody else presumably knows about the back door, so it hasn't been patched. And you can essentially use it. It's just an incredibly effective way to gain entry, usually into the whole operating system, for example, of a mobile device.

But these things apply to any kind of device. It's just an exploit that doesn't have any visibility whatsoever. Yeah, it's kind of like, you know, like every, there's this giant fortified castle and everybody goes in through like the back plumbing and they go through the pipes and then all of a sudden, they made it inside the keep and they're inside the castle. And do you operate in China? Are you allowed to operate in China? Do you even think about countries like that as a place to participate?

And how do you think about them? China is an amazingly huge market. Yeah. And, you know, it's the scale challenges for any business in China can just go from, you know, single monolithic node database, you know, two months later they need the biggest version of cockroach. So that's extremely exciting. And, you know, we are open source. And I think that's for us right now, the best way to really participate in the Chinese market. We've had a, we've worked a lot with Baidu.

And they use cockroach pretty heavily and provide it, you know, know, even to their customers in their private, in their sort of public cloud, which is like AWS or GCP. And, you know, that's a great way to work in China right now, you know, through a partner like that. It's not getting us much revenue at this point, but it's getting us exposure. And I think that's good. You know, I, I wouldn't say that I'm exactly optimistic that there'll be reasonable relations between China and the U.S., especially from sort of a business perspective. It's kind of hard to get Chinese companies to pay, but I think it's best, you know, where we currently are in terms of our scale to really think about how can we get

cockroach used by the most Chinese developers. And open source is a great way to do that, along with partners like Biden. Yeah, there's no way for you to stop somebody from using it, but that doesn't mean you have to, because your business is providing, I assume, support and hosting around the open source project, correct? That's right. Support tools. So they can put up their own servers and use it, and then they're contributing to the project. So you get that niceness of like, okay, they're scaling it. They're battle testing it in a world where, you know, their customer base is going to be, you know, I mean, the middle class in China, I believe, is bigger than the entire population of the United States now. We're close to it. So you just think about

that as a number. It's pretty amazing as an opportunity. But man, we have to reset relations with that country because it seemed like engagement was working. until it wasn't. Yeah, it seems that things are at the NADIR right now in terms of relations. And, you know, some of the human rights issues, listen, I don't want to get into big political discussion, but, you know, it seems that China's changed a lot in the last 10 years and not all of its positive. Some of it is. I think that there's always a chance to improve relations, and I'm hopeful that will happen.

It's just heartbreaking when you see, you know, something like Hong Kong fall, you know, and it doesn't, and we, and the world, at least the United States, it doesn't have the ability to participate in any way to say, we would rather that not happen. And is there a way we can work with you or incentivize you to maybe let people sell books there? And it's like, yeah, no, we make too much money. The movies make too much money over there. And, you know, the NBA makes too much money over there. And even you, from a modest company, you're like, wow, I got to be judicious and how I talk about this. I don't want to, you know, ruin opportunity there. And, you know, engagement has been great

for the people of China, that 300 million people coming into the middle class there, which is good for the world. I think, ultimately, it's great for the world, is because of the iPhone and because we're building stuff there. So we're going to do another podcast on that. But if you were to compare adversarial situations or challenging situations for you, which is more acute, dealing with authoritarian countries like China or dealing with juggernauts who want to take over every single aspect of open source like Amazon Web Services, which is a bigger threat to your business? Certainly Amazon at this point.

Amazon Web Services. Do they play fair? I mean, every company I've talked to who's in the open source space says that Amazon Web Services does not play fair. Why do people say that to me over and over and over again? Well, because in some respects, they don't play fair, but it depends on what you mean by fair. Amazon, I think, does live their company value, which is obsess about the customer. And that leads them down a different path. And for example, Google, I don't think they still have this as their guiding value, but it used to be, you know, don't do evil.

Yeah. And there's very different outcomes. if people hold that value, you know, relatively sacrosanct within the company. And people do. People adopt the values in companies. It's actually been a huge learning experience for me as a CEO. They, as long as your values really describe the tenor of the company and sort of the DNA, it persists and people adopt them.

And so I think from Amazon's perspective, they care about getting their customers, things that are less expensive. So they really care about that price point. and that are in terms of AWS at least, they are very integrated in a holistic set of services that can solve problems, especially for smaller companies that are, you know,

high growth, so the growth segment. And that makes sense that they're going to just take every open source product out there. Typically, AWS does what I'd call sort of good enough type offerings. Yeah. So it's like it's well integrated.

It's one sort of vendor throat to choke is the, the terminology people use. Yeah, it's like, What does it mean? Explain the metaphor. I mean, the problem is that I guess it was so, you know, as a, let's say you're a growing company and you have a bunch of different software that you use. If you have that spread amongst 10 different vendors, it's a lot of relationships that you have to manage.

Every year you're kind of renegotiating your price with each one of those vendors. With one throat to choke, it's like it all comes down to one vendor. So it's like you have one relationship, one person you're talking to, one billing system. The problem is I guess Amazon's throat is pretty difficult to choke. Well, yeah, and in fairness to them, the free market economics that they're doing is they're saying, take anything that's a cost center for Amazon, which their servers are a cost center, make them a profit center. This is probably one of the most brilliant moves in the history of business period. I mean, this would be the equivalent of, I don't know, Nike saying, we want to make our cost center.

of, I don't know, packaging sneakers and boxes. We want to make that a profit center. So package everybody else's sneakers. And, oh, by the way, stores that sell sneakers, we want to sell everybody else's stores and do third-party sellers. So it is wildly innovative. It's hard to understand, I think, as a consumer or as an entrepreneur, how to feel about it.

I certainly have mixed feelings about it because you do want to see prices go lower. As they drive prices lower with Amazon Web Services, they challenge companies like yours to lower prices, which then allows more startups to be born. And it makes you focus like a laser. If people are going to pick Cockroach DB versus the, does Amazon have a CockroachDB fork or competitor yet? Well, not that's public.

Not that's public in terms of a fork. They do have lots of competitive technologies. They have something called Aurora, which is probably the closest to what cockroach is offering. Do they try to poach your people? Are they like looking at your team and then just you wake up one day and ten people come to you and say, I got contacted by recruiters? Are they that kind of competitor where they just come in and try to to. That's happen.

In fact, they try to hire me sometimes. I get these emails from them. Oh, okay. So that's the blanket one. It's like, oh, hey, Mr. Kimball, we see that you have a passion for database projects. Where from Amazon Web Services would wondering if you would like to consider opportunities. That's exactly right. It's so hilarious.

I mean, I literally get emails for people and they're like, Mr. Calcutta, we see you're an executive in finance. We were wondering if you were looking for a financial partner to help you balance your portfolio. And I write them and I say, here's my book. Read the title. Would you, I think you should be asking if you can be an LPMI fund rather than if you should manage my money. Can you beat my returns? And I just throw them my returns.

And I'm like, okay, well, my first portfolio was a three-digit IRR. Can you beat that? And they're like, uh, no, her business is taking 25% of your four points that we make you in some blended bullshit. It's pretty hilarious. So what is your explicit strategy for dealing with a competitor who does not care about making a profit? Like Amazon Web Services. How do you rally the troops to be better, sharper, and what is it like?

So that's the internal question. And then the external question is when you come up head to head with them with, you said, Aurora is their product? Yes. They're not good at naming shit. That's the other thing is their names are kind of whack. I also find Amazon Web Services

is completely annoying. Like, this is my experience. They're like, hey, J-Cow, got a special for you. We'd like to offer all of your companies and we would like to give them all the stuff for like three months for free can you give us your database and i'm like yeah i got a podcast over here i have an event over here why don't you buy everybody at breakfast like we don't

have a budget for that and i say can i stop you right now you're you don't have a budget for that there are 18,000 people working at amazon web services i'm just making up a number here you you're making 20 billion dollars a year i literally banned a ws from coming to my events on principle because I would get literally 15 emails. They'd slide into my DMs. And they're like, hey, can you, we have a new Amazon Web Services product, JCal. Can you give it to all your companies? We'll give them all $1,000 credit.

Will you tweet this? And I'm like, what? Go support what I do in the world, Amazon Web Services. Like buy ads on the podcast if it matters to you. Buy everybody breakfast. Stop being so goddamn cheap. them speaking to the like 20 people I know over there, you guys used to, in the early days, they used to buy breakfast for everybody at my events.

And then for some reason, they're like, we don't believe in marketing. We believe in giving it. And then you know what they did to me even worse? You want to talk about insulting? This is when I told them all to just don't even email me anymore. And I told my sales team, don't bring me anything from Amazon Web Services unless it's a six-figure sponsorship deal. I don't want to even talk to this company anymore.

I'm so frustrated with them. they had the nerve, the nerve to give why combinator companies like 50% more than our companies and our accelerator. And I said, yeah, you know, I'm not going to even talk to you unless you match it. They're like, no, no, they do more companies than you. I said, you know what, guys, you're just insulting. And I'm tired of being insulted by you. So don't ever contact me again.

So this is a message to all Amazon Web Services, marketing people. ambassadors and evangelists. I don't want to hear from you anymore. Stop emailing me. I'm just so frustrated with you. You're so not supportive of what I do and the startup community. I'm tired of it. Period.

End of story. And I'm sorry to the sales team who has to deal with us now. But I'm team cockroach. How do you, so anyway, so how do you internally rally troops? And then how do you beat them in the marketplace? That's what I want to know,

because that's got to be like every board meeting you talk about that existential, you know, threat. Well. Sorry to go on a rant. Spencer. I'm just frustrated by it. I enjoyed it, honestly. It's easy to rally the troops internally because we're the ones that are actually doing the innovation. And that's why people at, that's why people join cockroach. They want to work on the cutting, the sort of bleeding edge of databases. And Amazon is repackaging other people's software. And they do a good job at that.

They're definitely the leaders. And I think they have a really good service in many ways. And obviously, you know, the innovation there is just astounding, especially the business model innovation. You know, so it's not hard to rally the troops. We don't really talk about AWS that much at board meetings because it's pretty obvious how we're going to build a business that co-exist with what they're doing, even if they were to Fort CockroachDB and offer it themselves. I think that what Amazon's really good at and where they have just definitely won the market and it's their incumbency in the cloud is around serving high growth companies. You know, they, again, it's like these smaller companies, they want one vendor throat to choke. They want one integrated billing thing.

They want a bunch of stuff that works well together and that's as inexpensive as it can be. But the real opportunity in the market is all of the world's biggest companies that have been for the last several decades on Oracle and on IBM DV2 and mainframes. Like these companies have all the money. Right. That high growth segment, it's amazing. It's huge. Amazon has killed it there. But what they're offering is very geared to these companies.

They explicitly look for ways to create friction so they can have lock-in into their cloud. This is not what a company that is now starting to embrace the cloud is looking for. They're very allergic to the idea of what they call vendor concentration risk. And Amazon represents massive vendor concentration risk, right? If you use a database with Amazon, as you say, it's like this 10-year kind of journey. That's actually pretty typical for a database. These big companies are, what they don't want to do is get stuck on AWS permanently. Yeah.

And the database feels like a good way to do that. And Amazon is very cognizant of that. And they like to push that high friction sort of integrated set of solutions. What we've discovered is that we're very good at playing for these, we call them the global 2000, Fortune 500, because what we're offering is something that meets them where they currently are on their journey to the cloud, which is going to continue over the next decade. But these companies have private data centers. They've got high value use cases. They have a hybrid future fundamentally for the next decade. They need to be able to straddle their private clouds to the public cloud.

When they're in the public cloud, they might start in AWS because that's where they have expertise in terms of their existing, you know, site reliability, engineers and DevOps people. But they don't want to be stuck there, right? Yeah. They want to make sure that they can. How easy is it? And is there a standard coming or under development, you know, from the Azure folks, Microsoft Azure. Am I pronouncing that correct? I've heard it multiple.

I've heard multiple too. Azure, I think, is the best way to say. Google Cloud. Is it Google Cloud Engine? GCP, Google Cloud Platform, yeah. I think they use a Google Engine. So GCP, then you have things like Rack Space. Is there an anti-AWS contingent working on portability and interoperability so that people can do best of breed and have it work outside of this Waldgarden kind of AWS issue? you? I don't see too much specifically from those other clouds, although I know that they have all kinds of different initiatives. Anthos is one that Google came out with recently, which is really about more of the private data center side, but I guess it's multi-cloud too. But there's a lot of companies. I invested one called Upbound or, yeah, is that thing that's one's called. But they're building like a sort of a multi-cloud control plane so that you can launch services in any one of the clouds. They're really abstracting. Wow. Yeah. Yeah. That's the, isn't that the holy grail?

If because I was on the board of a company called Dyn, the DNS routing company. And a topic that came up from a lot of customers was this ability to take their DNS and say, hey, send some companies to AWS and some here. I mean, that doesn't really exist in the world yet, does it? The ability to kind of route people dynamically to different cloud centers, is that just too hard to maintain for people too complex? Or is that what, like, the, company you're talking about that you angel invested in going to do? Well, yeah, it depends on what you mean by customers. Certainly if you have your service, and we have customers that do this, where they actually, they don't want to have their service fail just because there's a problem

with a particular cloud. And so they can actually put copies of the data on, say, all three cloud providers, say two, like, say GCP and AWS, and then they might have a private data center and they connect them with private fiber. And what they're doing is they're constantly managing copies of the data between all three of these, including copies of their execution software, so their application servers, right, that actually do the business logic. So what they can do is, let's say that Amazon misconfigures DNS in Europe and nobody can get to their data centers, right? And in that particular case, you would, you know, with your DNS configuration, you could have those customers fail over to using the GCP data center, the private data

center. And the way Cockroach works is it manages all those copies in a way that's consistent so that even if AWS is no longer reachable, you'll have exact business continuity, no loss of data, and you're now relying on the copies that existed in the GCP and the private data center. So it's a way to eliminate the systemic risk that comes with a cloud vendor, you know, misconfiguring or having some bug or some big outage or something like that. If we fast forward in 10 years, what do you think the cloud market share starts to look like because there isn't tremendous amount of money being invested here, a tremendous amount of development,

a lot of open source and some of this dynamic routing like you're talking about and people having stuff live in different places. There's also this concept of serverless where, you know, things fire up. Maybe you could speak to serverless computing if that is going to be a paradigm shift. Oh, yeah. What will this look like in 10?

So I'm going to break it into two questions. Talk about serverless, explain to the audience what that is and why that's important to people. And then explain what, if we fast forward, because you started on this journey in the aughts. What do we look like in 2035? I'm just going to pick 15 years from now. Wow.

I mean, that's a lot of prognostication there. Yeah, let's go. Certainly it's become immeasurably larger. I mean, it's not just that everyone's moving onto the cloud, and I was mentioning that as Global 2000, that tipping points already happened in the last couple of years. Now they're like every one of these companies is buying 100 million, or 100 billion in credits over the next four years or something.

You're like tripping over those kinds of deals. So that is already happening and that just represents non-organic growth in the cloud market. I mean, it's going to be getting up there in the hundreds of billions and maybe even close to the trillion dollar mark over the next 10 years. Crazy. But it's also, it's not just those companies inorganically moving their spend to the cloud. It's also that you can do things faster with the cloud. And that's really where serverless comes in.

You can iterate faster. You can deliver more services, more applications to your end users. that's how you compete whatever vertical you're in. Let's say you're in financial services. Like it used to take, you know, months just to get machines into some private data center that you could use for something you were going to launch. This is months. Now you can get those things in minutes, maybe even seconds,

especially when you're using orchestration technologies like Kubernetes and things like that. What serverless fundamentally is is it's subtracting all of the deployment, all the work that has to be done around deployment. from the equation. And they're saying, okay, we're going to do this deployment DevOps, monitoring, keeping all your services up, maybe some auto-scaling type stuff. We're going to do all that for you.

And all you're going to do is pay us, right? And so the total cost of ownership is going to be lower. You might be paying us a bit more than if you did it yourself, but that's not counting all of the expertise that you have to hire into your company to run that yourself. So if I were to put this in a model to explain it to somebody in the most basic way, If I wanted to put up a website, a commerce site, I'm selling sneakers.

I'm like Zappos 1.0, which is very basic. I'm a sneaker company. In the old model, I buy a server. I buy a T1. I put it in my closet. Or then it was like I put it, I co-locate in a data center. Then it became somebody who was a managed service like Rackspace. Then it became the cloud like AWS and Rackspace and other things.

And now what you're saying is, hey, here's a little bit of code. So when you load the website, that serverless code fires up and then show. shows the website. And then it shuts that service down. So if nobody was loading the website, there's no dedicated hardware somewhere just waiting for those users. And if the sneaker goes supernova because some Kardashian wears them and a 10 million people go to it in the same day, that same little code executes, but goes out across any number of servers and service them. Correct? Is that a simple explanation? That's a very good one. I mean, the horizon that I'm looking at the 2035 question.

I think it's actually going to happen more like by 2027 or so, is that, you know, developers can develop a, you know, mobile app, web application, all on their laptop, right? They have a local database and they're doing all the development. But then what they want to do is they want to push that into the cloud. And that's the last they want to deal with that, except for paying for it to scale globally.

Yeah. Right. So they, in other words, like, you might launch your new service and nobody gives a crap, right? You haven't done your marketing yet. no one's really using it. Maybe your developers are poking at it and make sure it works and you get like 10 requests a day. You're paying for 10 requests a day. You're paying for just that little bit of storage they might be storing into serverless databases, right? You're only paying for application servers because it's like a Lambda or GCF function that pops up just like you say to handle that one request or those 10 requests per day.

So you're paying like cents over the course of a month to run this thing. So crazy paradigm shift for people who are just got some, you know, $6,000 a month AWS bill and they don't know what they're paying for when they figure out the serverless thing exists and they could be paying 60 cents or $6 or $60 or whatever it is. Right. And that would just scale up, right? And you want that to scale up so that it can go maybe you're paying literally $600,000 a month because it is scaled globally and you're handling hundreds of billions of requests a day.

But you're also making, you know, $6 million a month. So yeah, you're like, okay, I'm happy to pay for this. I mean, there's lots of stuff that's unresolved in terms of how that happens. but fundamentally you want to say as a developer, I want to take what's on my laptop. I want you to scale this globally and I want to pay only for what I use. So if it starts off as being nothing,

then I pay nothing. If it goes, you know, balls to the wall and everyone in the entire world is using this, I'll pay more, but you're going to do all that scaling. I'm not going to have to re-platform. It's exactly what I gave you that worked on my laptop. Now it's going to work everywhere around the world. That's the holy grail.

That's the dream. And that dream occurs if we were to pick and you ever do gambling? You ever gamble? Set a line? You don't gamble. Okay. So setting the line is you set the line I picked the over or under.

So what would you say is the year that you think this becomes the default for startups? Just pick a category startup. Startups just all go serverless. What's the year that happens? You had to pick a year, 20s. It's never all going to go serverless. But you're talking about like that. Okay.

Let's say the majority of them. Yeah, yeah. You know, or like some, actually, majority is kind of silly because you have all the existing stuff. But the majority of new startup, so you take a Y Combinator Techstars, launch accelerator, those 2,000 companies, the majority of them go with a serverless paradigm, just for that segment, because that is the future of us. Yeah, I mean, I'm pretty confident that it's going to happen by 2030.

But, you know, I'm optimistic it'll happen before then. Got it. So then you're taking, you might set the line at 230, and if you had to bet your entire net worth on it, you'd say under 2030. Entire net worth. Yeah, that's how we gamble. I'm not holding you to it, but it just so it's, kind of fun ways to ask questions. Hey, when we were, as we're wrapping up here, I want to, and thank you for the time.

You know, it's, it's, it's, it's really challenging to have a guest like you'll be honest, because there's so much to talk to you about. But, you know, we're doing so good on thinking about tech in a way that is accessible. I want to talk about the paradigm shifts that are coming. And, you know, like, I'm talking big stuff. So it seems to me, you know, serverless is one of those paradigm shifting technologies. quantum computing is one that's often cited. Power and power consumption and then CPUs generally, storage is sometimes a blocker. When you look at things that could shift paradigms in the next 50 years, and now we're now going to sort of go from the midterm to the long term here, what are the paradigm shifts that our kids and their kids will be looking at,

when it comes to computing? Well, one big one that's going to happen in the next several years that you didn't mention would be 5G. I do think that it's rare to get latency, significant improvements in latency. That happens very infrequently in computing. And I think that there's, you know, the real-time experience is something that most people just haven't really imagined

what's possible. Okay, so let's stop for a second and explain. latency as it currently stands now when I'm using my cable modem or my 4G phone LTE versus what it would look like there and then what that experience would open up of extremely low latency. What would we be talking about here? So when you really think about latency, for most applications, obviously high frequency trading and some gaming is different, but mostly what you wanted to do is get under 100 milliseconds. That 100 milliseconds is kind of like, I guess the U.S. Department of Defense figured this out.

this is the command and control system, um, limit threshold. If it's under 100 milliseconds, it feels instantaneous to a human being. Got it. If it's over 100 milliseconds, you've noticed the lag. And so, uh, for certain things like, which is a tenth of a second for people to understand, correct? Exactly. 10th of a second.

So it's not, it doesn't seem like much, but it's an eternity in, in computer world. But like, you know, the reality is that when you, everyone on their mobile phone right now hits buttons and they are, they find it extremely normal to wait a second or more. for something to come back. Right. And what could happen is, you know, 5G is very much an enabling technology because it's basically going to ensure that, you know, for all of the users out there that have 5G, you're able to get on the backbone within, say, you know, around 10 milliseconds,

10 to 20 milliseconds. And that's, you know, that's kind of true with LTE right now, but it's not, it's not very consistent. But the combination of 5G and also startup companies and bigger companies, you know, anyone that has a service, they have to actually co-locate the application servers and the data next to the customer. That's why when I was talking about this horizon where you squint at it and you say, here's what serverless could be. I'm talking about scaling globally. And really what that means is giving global customers a local experience, which means that you need the 5G so they can get

close, they can get on the backbone quickly. But then you need to make sure that the application and the data is stored close to the customers. So you can't have an Australian user hopping across of Virginia. And that is going to be like a really big paradigm shift when you can actually easily democratize the ability of even a startup to create a global application data architecture so that they can give customers wherever they show up. Let's say they show up in Brazil. You don't really know when you're a startup or Tokyo or something like that. You want to give all of those customers a wonderful sort of first class user experience. And I think getting everything under 100 milliseconds in your application is going to enable a lot of what feels very fundamentally

different in terms of how applications behave. Like imagine when you're sending a tweet out or you're typing something. Right now you might get some little... Yeah, a little spinning wheel of dev, little dot dot, dot, ellipsis, whatever. So is typing. Yeah. Exactly. And what you could do instead is really create a vibrancy through, you know, real-time interaction.

with the larger community. That's going to change how things feel. You'll feel this connection, which right now, you know, there's like this, everyone spends so much time you just walk out on the street, I mean, less than COVID times, but everyone's sitting there glued to their phone. We're spending hours a day, like through this interface into these virtual worlds, but it's very slow. So the interface is like this incredible lag, which doesn't feel so bad compared to writing a letter

or, you know, sending an email even. It feels quite real time compared to that. but it could feel more like real life. And I think that's going to be a, that's going to usher in a lot of new use cases, a lot of change and a lot of opportunity fundamentally. Yeah, and the, it has to be the double punch of, hey, I got this low latency.

So I requested my Gmail. But if I'm in Taiwan or I happen to be on vacation in Sydney, and my Gmail is in the United States, well, that doesn't help me because now it's got to go across some backbone and it defeats the purpose of it, which I think is one of the things I heard is that some services, if they see that you're traveling, will then move your database, let's say an email database, or if you were a customer who was doing databases, and I, you know, as my e-commerce company, my Shopify store might then be co-located via what's called a content delivery. work or CDN to those endpoints. Is that correct? Yeah, there's lots of, lots of different ways to do

it. And it becomes potentially pretty complex in some case. Like someone was just traveling for a day through Paris and then at a day in Barcelona, you know, kind of having their Gmail follow them would be a lot of data transfers. You wouldn't necessarily do that. But if they moved, you'd absolutely want to do that. Does that happen dynamically? Do you think, like just without putting Google aside, but all email servers, you think they actually do that kind of stuff? The ad scale ones? But when I was at Google, Gmail didn't automatically move what they call your home if you move to Europe or something like that in the United States. It might happen automatically now, which would be pretty sophisticated. That's a sophisticated move, yeah.

Yeah. But certainly for something like, I don't know, Disney Plus, if I'm pulling up the Mandalorian, I am not pulling up the Mandalorian from some data center in Canada. I'm pulling it from some node that's very close to my house, correct? Exactly right. I mean, it's extremely expensive for them to deliver that over, say, a transatlantic link or something like that. So, yeah, I mean, Netflix does this too. I mean, everyone that's actually delivering, especially read-only content, that's a more solved problem. The really interesting stuff is, like, what do you do for things that have to be read and written? Like your Uber account, for example. So Uber has a global data architecture, and they probably spent, you know, 100 engineering years on it, like literally a century of someone's time, but there's obviously a lot of people working on in parallel. but that allows them to have a holistic global service. So you have Uber no matter where you are and it works well no matter where you are. It might not work as well when you're traveling in terms of you might have a little bit more latency as your account is probably stored mostly in the United States.

But certainly if you live in Australia or Sydney, let's say, you're going to have a, you know, from day one, you're going to have an amazing experience even though you're an Australian user. And that's what everyone wants to build those kinds of data architectures going forward. And I think that's, you know, something that, you know, Cockroach is an integral part of that, and that's some of our functionality that really does enable that. But there's lots of other pieces, the architectural stack that are going to be need to be put into place and integrated so that even a startup is able to get that kind of functionality. Yeah, it's really going to be interesting also when Starlink and some of these low Earth orbit satellites get up there. And they have 100 millisecond or less time.

You're going to have 5G in dense places and then, you know, everywhere is going to be blanketed. You're going to have this like double whammy in the next two or three years, the idea and the competition that's going to come to that, you know, last mile or last whatever number of miles it is to a satellite. Just anywhere anybody wants to work. And this is going to propel this work from home concept, I think, to another level. Because right now people are buying homes based on the Internet speed. Like it's literally in the description of the home. Oh, you want to move up to, you know, Saugaties or, you know, Hastings on the Hudson. great, you know, but what's the internet in that town like? And any other paradigms that you follow

like really closely and think are just going to be transcendent? Like, I don't know if you follow quantum computing, which was supposed to be here a million times. Quantum computing is very interesting. I mean, you know, that, yeah, I don't exactly know when to expect that. But I do think stronger and stronger forms of, you know, sort of general artificial intelligence are a real possibility. You mentioned our children's lifetimes. I think that's, you know, I think that's, that's, you know, probably somewhat realistic that some of that will start to happen. You know, think about machines being, or you say artificial intelligence as being better at virtually every meaningful task than humans.

I mean, that's going to be such a paradigm shift. It will be probably politically destabilizing in terms of just the nature of human work and capitalism. Yeah, or it could create just such a hyper-capitalism that we've never been. ever seen before. I was thinking about the other day. I don't know if you want, is it GPT3 or GTP3? I can't remember. I'm not sure. You know, the, the open AI did the thing where you can just put in words and it talks to you and it kind of tries to talk in your voice and it's like, I haven't seen this. Oh, yeah. It's anyway, it just became like, well, the rage. But I mean, there's literally

going to be a moment in time talking about movies and the creation of films where you and I could be on a Zoom call and be like, hey, let's make a movie that's like a Star Wars movie. And, Okay, let's have Bo Bethet fighting Obi-Wan Kenobi, and then let's add in you and McGregor's Obi-Wan and a flashback to do this. And then let's have him say something really witty to Han Solo. And like literally, boom, it's going to be like a perfect rendering that looks no different than the Star Wars that we've watched over the last 30 years. And instead of it being fan fiction, because you could do that right now by talking into

voice recognition technology. You could do that in some number of hours. People are making fan films for Star Wars. I don't have you ever seen them where the lightsabers look better than they did in the original Star Wars trilogy. But being able to then recreate the actors and say, you know what? Let's recast Darth Vader with, you know, this voice. And it'll be like, boom, here's the new version for you. We ripped it.

And it'll just be sent globally. I mean, that's literally going to happen in 30 or 40 years. I would like to see Jar Jar-Jar Banks is like one of these evil Jedi's. What are they called? Sith, yeah. Let's make Jha Jha. Sith. Oh, it exists right now in the fan fiction world.

I mean, and there's probably some darker things that exist than just. Actually, there is, if you want a little, are you a Star Wars fan at all? Yeah, I am. Okay. Have you watched the Clone Wars by chance? You have. This is the big secret that nobody knows about. I didn't know about this until I was talking to John Favro, you know, the director who did the Mandalorian.

And we were in a VIP room, and I was talking to him about the Mandalorians. I kid you're not. And he was like, well, you know, have you watched the Clone Wars? I was like, no. He's like, well, you know, there's a whole thing in the Clone Wars about the Black Saber. There's a whole thing about the Mandelaar. You got to watch that episode. And, you know, that's part of the arc here of what's happening in the Mandalorian.

And I was like, oh, that's cool. I was like, man, Disney is so lucky. Like, man, they must have backed up the brinkstruck or whatever. And he was like, Jason, I had to go in and pitch them like seven times over three years to get them to let me do this. I was like, but you're fucking John Favro. Like, shouldn't they just a second you have idea and do it? He's like, that's not how it works in this industry. That's your industry.

And have you watched Rebels yet, the follow-up to it? No, I haven't. I just, I'm almost finished with Rebels. What I do is to get my Star Wars fix and to stop being, you know, my quarantine 15, I just go on my treadmill and it's 22-minute episodes and I watch it and it's just like my childhood all over again. All right, listen, Spence, you've been a great, great guest. and continued success. I could talk to you for hours,

and we talked for an hour and a half. So it's GPT3 anyway. Yeah, it's like a little AI thing. I'll check it out. Yeah, it's just basically everybody went nuts for like 15 minutes over here

because it literally can write in your voice and you put in some data set and all of a sudden it starts talking. And, you know, it's sort of like, do you have the Gmail feature where it starts giving you like two words ahead? So you're like, I don't think. And then it's like, we should.

And you're like, okay. write arrokey we should sure why not and i'm like writing this as a writer and i'm like i don't like this i don't want you to leave the witness it's like leading the and if it's going to lead the witness like i want to be able to pick like hammingway or joys like can we lead the witness and and make me sound not like everybody i bet that's the next the next innovation there is you can pick the author that you want to channel and then the authors can get a license so you pay a penny per word to write like them wow that's a that's a good that's a business model i like that coming in twenty twenty seven from amazon web services uh all right spencer kimball cockroach labs dot com if you're

looking to work for a great company not the evil empire i'm not picking any company that's the evil empire here. I'm just talking in general. If you want to work for the tip of the spear and do innovative shit, hey, Cockroach, DB, Cockroachlevs.com. And I assume you're all remote now. Has that been a big transition for you? It's been a good one. We slipped into it pretty easily. We were about 30% remote before. And, you know, the great thing is this had happened 10 years ago. All the video conferencing stuff wouldn't be up to snuff, but it's pretty good these days. It's pretty amazing, like how effortless this has become. I was never a fan of it as an invest.

We didn't even get into investors. I got to have you back on the pod just to talk about investors. But as an investor now, it's like, I can't invest in somebody unless I'm in the room with them. I got to break bread. I just got to go for a walk with them. I got to do a walk and talk. I got to have an espresso.

It's got to be like two or three hours in person at least if I'm going to cut a check. And now it's like, I just got on a 15 minute Zoom. My team's pre-vetted them. I'm like, here's 100 grand. See you on Zoom. And I'm just like, I just, I just, I literally have invested in 40 companies during pandemic before $4 million to work in 40 companies, 100K each. Wow.

And just like, I'm like, whatever, if this is the new world and nobody else wants to make these bets, I'm going to be, I'm going to make the bets. And for everything you give up, you add something, you know? Yeah. What is it for you that's, this is added to you as a leader of a company? Have you thought about that at all? Like how you've become better? I think the, well, I mean, as a company, I think the big opportunity is to hire from a much bigger talent pool. We used to be, you know, just, Toronto is kind of new for us, but San Francisco Bay Area and New York City.

and now it's yeah the sky's the limit you know it's basically where do we already have business entities and you kind of want to stick to those because you don't want to add too much overhead but I think that's that's a big that's a big if you don't as a company like ours rise to meet this challenge and the inherent opportunities I think you you put yourself at a big disadvantage yeah every crisis is an opportunity chaos is a ladder I mean there's a million different ways this has been phrased but it's really I'm just so impressed the entrepreneurs in my portfolio and how they fought and really moved on a dime. And I really wish we would see that in some other aspects of our society where people maybe could be a little

more nimble in addressing things like what's going to happen with our kids in three weeks or two weeks or next week for school. I don't want to get into it too much here. But my lord, has our government and our systems failed us? And we need to be entrepreneurial. And that's what solves problems is, you know, people taking risk, right? And taking chances. and maybe we're doing none of that. It's just so impressive. Outthinking this status quo, right? And I think that's where the innovation from entrepreneurial energy, that's where

it's best invested. You got kids? May I ask? I have a three-year-old daughter. Okay, so nursery school or stay home, I guess, or possibly? These days. Yeah, I think this year we're probably, I mean, her school closed and the school closed permanently. I can find another one.

but I mean, at three years old, really she just needs to play. I mean, she was going to Montessori school. Oh, me too. I'm going to have four-year-olds in Montessori's. I'm watching them like cut vegetables. Like, have you got them cutting the vegetables yet with the safe knives? I think they might have done that. I didn't see them do that.

They arranged flowers. It's really, it's a cute little program. Here's where you do. Okay, I'm sorry, finish your sentence. No, I was just saying, I think she was doing well there, but honestly, at three years old, it's not like I'm going to insist she goes back in the middle of COVID. I mean, I went to public school in Brooklyn, and they were just, just like, sit down and shut up.

And then they were like, this is the third time we're going to tell you, Mr. Calacanis, sit down and shut up. Like it was like literally, I thought my name was sit down and shut up. Because that's old I ever, he just never told me was shut up. I think I was the last generation of people to get hit.

I remember distinctly in high school getting smacked on the back of my head by brother Thomas. I'd never forget it. He just cracked me. And I was like, it's not allowed? He's like, eh, maybe not.

Maybe, maybe not. But I mean, Montessori, if you just go on Amazon and type in like Montessori stuff, they're very big on whatever the child can do, having them do for themselves. Whether this is putting on a pull-up diaper or cleaning their plates. So my wife introduced me to Reggio and Montessori, and I started reading about the stuff online and trying to catch up. And now what I do is I'm like, okay, we're having carrots. and I put out two cutting boards, they're twins, and I have these special knives that they, you know, if they were to slam it on their finger,

it might hurt, but it's not going to slice them open and we're not going to the emergency room. And they just sit there and they cut the carrots up and put them in a bowl. And I've got to rinse them. Okay, now I want you to dump them in the steamer. And I'm just slowly doing this. And then, you know, on a weekend,

I'll wake up my 10-year-old has made pancakes. And I'm like, whoa, here we go. And it's that like resiliency, because you go to Burning Man as well sometimes. Yeah. And, you know, I tweeted, the other day, I was like, I'm a fan of radical self-reliance. And that is one of the key tenets of Burning Man.

And I have never gotten so much shit on Twitter. I mean, I have. I've said some stupid stuff. Kind of like Twitter is, yeah, it's dangerous for people who like to talk. But I think that's what the defining, because we're going to solve this thing. And I think we'll solve it by Q1 because we'll have Biden and Harris and they'll just shut the country down for six weeks and we done. And we'll get through it, national mandate, which we could have done in March and April, this would be over like everywhere else. But the radical self-reliance that people picked up here that, oh, I can leave the city. I can go find another home. I can put chickens in my backyard and take the eggs and eat those. I can stock up on this. I can I can educate my kid at home, whatever it is. I can run my company even, you know, if the company is or, or my. my restaurant could have street seating if it never had that before. That radical self-reliance, that's the silver lining, the entrepreneurial creativity and

self-reliance, which is what Burning Man is about, which is why Burning Man is so easy for people to be cynical about. And for an East Coast, I was like, yeah, I'm going to go to Burning Man. I love going to Burning Man. The art, the people, the costumes, the music, everything is fantastic. It's the energy. It's like, the energy is palpable there. Yeah.

When did you first go, by the way? I guess it was seven years ago. Yeah, I think the first time I went was like 2005. Oh, wow. So you're a long time. And there were 15,000 people there. And you know what they told me? Don't go.

It's over. It's ruined. It's too big. It's over. And I went like two years ago. There were 70,000 people there. And I was like, this is 10 times better. The year I went, there was just the circle.

And it went two streets back, maybe three. three. Oh, wow, that's it. That's it. And the circle, there were like three neon signs on the entire circle. That was the complete lighting of the place. So you're like, okay, that's the disorient camp. Okay, that's this camp and that's that camp. Okay, I know where I am. Now you go and the entire, I mean, you see the Mayan truck? Yeah, the Mayan Warrior. The Mayan Warrior truck. I talked to the guy. It's a $10 million. It's a $5 million laser show and you can, that's what I want for my birthday. Anybody, I'm going to be 50 this year. Can somebody get me the Mayan, what are they called? The Mayan warrior. The Mayan warrior. I also like the other one. What's the love? The heart.

Oh, that's the robot heart. I'm like robot heart. I like the Mayan warrior. What do you? What do you like? I like both of those. I think that, I mean, listen, Mayan warrior can't beat that light. Can't be it. Can't be it. Yeah. What about, what's the one that happens during the day? There's one that's like the daytime one. You mean an art car? No, not an arc. There's a dance floor that's during the day, a stage that everybody goes to. It's like if you want to dance at like 11 to 5. Oh, it's that, well, I don't.

It's a pretty good one. It's a San Francisco one that's really dope. I forgot the name of it. Everybody always meets there. They always have the sunset there. It's over on like 9 o'clock or something. It's not, and there's the, like, the Cirque de Soleil, not them, but there's like a whole trapeze next to it. And my best moment of Burtigman was, I go by the sign and it says like, you know, rule number one, rule number two and like safety third.

And I was like, okay, safety third. Great. There's a message for our children. What's your best moment, Birdingman? You got a favorite, favorite moment, a favorite thing to do, go to the temple or? I love the temple. A lot of people don't. I find the temple.

It's pretty hardcore, yeah. It is hardcore, but, you know, all the sadness there is the flip side of all the happen. that generated the sadness, right? It's like this two sides to a coin. I find it somewhat beautiful, and I've had lots of, I think, kind of transcended experiences at Burning Man in general and certainly at the temple. I think my best thing was just my first year, the first day I was there. I was like a kid in an amusement part for the first time.

You know, I was like, I haven't felt that way since I was 10 years old. And to feel it again at when I was probably like 41, it was pretty mind-blown. I was like, oh, my God, I remember this feeling. It's like, this is the feeling of just, yeah, excitement. Mind blown. But that was the mind too. And I was a pretty cynical guy about it.

And I was like, I'll go, whatever, hippie-dippy nonsense. And I was like, this is the greatest art music. People are in the right mindset. And, you know,

when I say about it, my friend Dave Goldberg who passed away, you know, that temple, I was like one of the most deep experiences I've had in my life

because I just sat there with other people who were grieving people they lost the year before people don't know that there's a temple and then there's the burning man. They burn the man on Saturday night.

They burn the temple on Sunday night. And people go, and they put a note or something they remember about a person who died in the previous year. So you go, and there might be, what, 500 people, a thousand people around the temple, in the temple, grieving, their loss. And it's just, it's one of the most beautiful things I've seen in my life. It's very spiritual.

And then they burn it, which is a really interesting, you know, everything's ephemeral. And sometimes people hold on for these things too long, but you can invest a lot of your, I think, grief and pain in this and then having it all burn. It's very somber, but it is beautiful. It's sort of a renewal. Yeah, you know, you put the, yeah,

I see people will put up a picture and then they'll write the story. And then you walk around it and you get to see like, oh, yes, we are on this planet, but for a moment. Make it count, right? Make it count. Yeah. All right, Spencer, listen, this has been deep.

I will see you 2021 at the burn. Hopefully if we get this done. Hopefully. I think we're going to get it done. If we can get a little bit of change in the management of our, you know, of the nation. I'm hoping for a bigger change.

Yeah. Maybe we could get somebody in there who would take responsibility. Maybe people could pick up a couple of these. Adam sells these beautiful masks and they give one for free to some people. You wear it and then it goes away. I don't know. All right, listen, be safe, Spence. And great to have met you and great job on the pod, by the way. We'll see you all. Thanks, Jason. It was a real pleasure.

Real pleasure for me too. Okay, cheers now. Bye.