What We’re Reading (Week Ending 21 November 2021) - 21 Nov 2021
Reading helps us learn about the world and it is a really important aspect of investing. The legendary Charlie Munger even goes so far as to say that “I don’t think you can get to be a really good investor over a broad range without doing a massive amount of reading.” We (the co-founders of Compounder Fund) read widely across a range of topics, including investing, business, technology, and the world in general. We want to regularly share the best articles we’ve come across recently. Here they are (for the week ending 21 November 2021):
1. Experts From A World That No Longer Exists – Morgan Housel
Henry Ford was a tinkerer. He revolutionized the factory floor by letting his workers experiment, trying anything they could think of to make production more efficient.
There was just one rule, a quirk that seemed crazy but was vital to the company’s success: No one could keep a record of the factory experiments that were tried and failed.
Ford wrote in his book My Life and Work:
I am not particularly anxious for the men to remember what someone else has tried to do in the past, for then we might quickly accumulate far too many things that could not be done.
That is one of the troubles with extensive records. If you keep on recording all of your failures you will shortly have a list showing that there is nothing left for you to try – whereas it by no means follows because one man has failed in a certain method that another man will not succeed.
That was Ford’s experience. “We get some of our best results from letting fools rush in where angels fear to tread.” He wrote:
Hardly a week passes without some improvement being made somewhere in machine or process, and sometimes this is made in defiance of what is called “the best shop practice.”
They told us we could not cast gray iron by our endless chain method and I believe there is a record of failures. But we are doing it. The man who carried through our work either did not know or paid no attention to the previous figures … a record of failures – particularly if it is a dignified and well-authenticated record – deters a young man from trying … I cannot discover that any one knows enough about anything on this earth definitely to say what is and what is not possible.
The important thing is that when something that previously didn’t work suddenly does, it doesn’t necessarily mean the people who tried it first were wrong. It usually means other parts of the system have evolved in a way that allows what was once impossible to now become practical…
…“Don’t buy stocks when the P/E ratio is over 20” was a good lesson to learn from the 1970s when interest rates were 7%, the Fed hadn’t yet learned what it’s capable of, and most businesses were cyclical manufacturing companies vs. asset-light digital services. Is it relevant today? At a broad, philosophical level, yes. In practical terms, probably not. In the same sense, buying stocks at all seemed like nothing but speculation in the 1920s because corporate disclosures were so opaque. By the 1970s that had changed, and you could begin to make rational, calculated long-term decisions that put the odds in your favor.
What was foolish to one generation was smart to the next, but the older generation’s views lag. Every generation goes through this. Every generation fights it.
Same thing in the economy. Take this simple change in how the government views stimulus:
“Liquidate labor, liquidate stocks, liquidate real estate. Purge the rottenness out of the system.” – Treasury Secretary Andrew Mellon, 1930
“We have a lot of money. We need to get that money in Americans’ hands.” – Treasury Secretary Steve Mnuchin, 2020
That’s an enormous shift, an evolution in how policymakers handle recessions. Investor Conor Sen recently pointed out that high stock valuations and low interest rates used to mean future investment returns would be low. But now, he wrote:
What it’s actually indicating is that there exists a lot of fiscal capacity for higher levels of government spending, which can boost real GDP and earnings growth (probably some inflation too), an outcome better for financial markets than you’d get without that policy shift.
But for older investors whose careers have overlapped with high inflation and a policy environment dominated by monetary policy they’re not thinking about this — it’s the flaw in their framework.
Not a day goes by that I don’t become more confident that the secret to business and investing is identifying the few things that never change and hold onto them for dear life, and identifying what evolves and be ready to adapt those views quickly.
It’s just so hard to do the latter.
2. Eric Golden – Bored Ape Yacht Club – Patrick O’Shaughnessy and Eric Golden
[00:03:39] Patrick: And if you think about the collection, so there’s 10,000 of them. They’re NFTs. People collect one or more of them. What are the motivations for owning one that are distinct from let’s say a pure piece of art that you might show people but doesn’t have a membership criteria to it. Like how is this something more, something different?
[00:03:56] Eric: So if you zoom out, I think for me of how I got into this, this is not what I thought I would be owning. I got into crypto late. And by late, I mean everyone who was after 2013, that seems to be a big demarcation. And so I was into crypto in 2018, 2019, and I was buying it to coins. I really didn’t know much about it. But went on Twitter, most of the information was coming from Twitter and Discord. And on Twitter you would see these CryptoPunks, and I really wanted a CryptoPunk, and you would go in their Discord, and I was lusting after these things. I was looking at them and saying like, “I just want one.” But when I went into the Discord and I met these people, they’d been in Bitcoin since like 2010, they own Glitz. They just knew everything. And I felt like such an imposter. I was like, “I’m new to space. I’m just trying to get skin in the game to understand it.” And I really didn’t feel like I would belong.
And so when the Bored Ape project happened, it had come right after Top Shot. So Top Shot came out at the end of ’19. I got big into that in the beginning of 2020, and people were trading basketball cards, and it really was this first experience with NFTs and Roham Gharegozlou, the founder of Dapper Labs, gets a huge amount of credit of onboarding us into what is an NFT.
And so at that point, ironically in April, Larva Labs was about to issue their second project called Meebits, and I actually bought two of those. And I thought it was the most ridiculous thing I could ever do. I minted a human and a pig. A pig was very rare, and I showed my wife and I’m like, look, we just minted a digital cartoon and parents of a new pig. And I’m like, you can trade this thing for like $30,000. This whole thing crazy. And I put it up as my avatar, and I was trying to make this personality and just play around in the space. And the Apes were like taunting me. And they were this new project. They started off at $200. Meebits came out at like the equivalent of $8,000.
And so this was kind of a blue collar project that it had a vibe and a culture I wanted to be a part of. And so I ended up buying some Bored Apes and joining the group and joining the Discord and learning more about it. And I would say for a lot of people, especially I bought it in May, so it was a one month after launch. It was kind of this silly thing, which was like, we wanted to be punks, but we weren’t punks. And we bought this picture of an ape, and it was this fun lore of what would happen if everyone got rich from investing in crypto, what you end up doing, you would create this really cool bar and club to go hang out with your friends, which resonated with a lot of people. Like you’re working really hard. You want to make a lot of money, but what’s it all for? It’s just to go hang out with your friends.
And so this culture war started between Meebits and Apes, and we were joking about it, what group you were going to join. It was all very playful and fun. We were watching the price of their project, which is quoted in floors. So floor is the cheapest you can buy something at. There’s was two ETH, and ours was 0.3ETH, and we’re like, “Someday we’re going to cost more than Meebits.” And we would just joke about it in the Discord and talk about what this project could be. And so it really wasn’t on buying a piece of art. I think this is beautiful. I want to hang it on my wall, although I did really like it. It was more symbolic of joining a community and having a shared belief of the people who were on the frontier that are willing to fail unconventionally, but came into crypto around the same time.
We weren’t the CryptoPunks. We weren’t the first to get there, but we were totally okay with that. We were this new group that really wanted to have fun and teach each other. There’s another famous podcast that always asks what’s the kindest thing. And I really felt like the Bored Ape Yacht Club exemplified that. When you went into the Discord, you’re like, “I don’t know what DeFi is. What is Solana and why do all the VCs own it?” It was just this amazing sharing community where with my ape, I could speak in a way that immediately got me into a circle of people I wanted to associated with and I wanted to learn from. And it was really a club, not a piece of art where I was going to say, I own this rare thing. I’m going to hang on my wall…
…[00:11:29] Patrick: Let’s talk now about the value of holding these things beyond the purely digital. They release this thing called Roadmap 2.0, and again, I want to come back to, there’s a difference between owning a piece of art, say in the extreme you own the Mona Lisa, that’s incredibly valuable. Maybe it appreciates over time. It’s a piece of culture. There’s all sorts of interesting art specific angles that we’ll probably cover more in other episodes. But in this one I’m really interested in sort of the membership and community specific angle. So how is this organically, this roadmap managed, how does the community contribute to it? Are you excited about it? What do you think this might become if each of these 10,000 things is effectively like a membership card into a club?
[00:12:11] Eric: Roadmap 1.0 was a scavenger hunt where we were going to give away an ape and also a certain amount of ETH. So the first Roadmap was pretty basic, but the big finale was releasing of the mutant apes, which really was a way of extending the club. So if there was 10,000 people that were part of the original, they had always planned to make that number in total 30,000. But was we really fascinating was the way they went about distributing the second batch of the 20,000. And what was interesting about it is most of the wealth of the money paid for those mutants went to original owners, and some went to the Bored Ape Yacht Club founders themselves.
So there was this economic thing where they delivered to original holders Serum, and that Serum, if you combined it with your ape, would create a new mutant ape, and then you could sell that. But you got that for free. The value of that the day it was dropped in U.S. dollars was between 15 and $60,000. So everyone who owned an ape just got 15 to $60,000. And that came on the heels of us already getting a Bored Ape dog, which we could have sold for probably 15 to 20,000.
And so if you were in early, you’ve been paid back your investment already through the airdrop mechanisms, which we can talk about some more. What the Bored Ape Yacht Club itself raised, the actual founders raised something like $90 million in that second sale. Combine that with the fact that one of the interesting things about how NFTs sell is every time a transaction happens, the Bored Ape Yacht Club gets 2.5 per percent and a royalty. So if you have a million dollar sale, you have $25,000 going into the kitty. And so these numbers are wild, but since April, they’ve done over a billion dollars in sales and raised over a hundred million dollars in revenue for this organization to do what it will with.
And so you asked the question, there’s kind of two parts. One thing that made Bored Apes unique was they let the owners take the IP. So I own my ape. I can start a beer company, a coffee company. We can start the Bored Investor Yacht Club together. There’s all of these things that you can do because you own your own ape. Roadmap 2.0 right now is literally just a picture teasing out a lot of different things. And so the first thing that’s coming out is called Ape Fest in New York, which I’ll be attending, where we literally rented a yacht to go out on the Hudson River for a thousand of our friends to hang out. A lot of us have never met each other. We’re also throwing a huge party in Brooklyn, and there’s a bunch of other activities happening for this week. So that’s kind of the first meetup in real life.
And then there’s teasing at things that people are guessing about or speculating. You can pull the picture up. One is a game, which a lot of people are talking about, like a game that group could play and things that we own, we have some sort of part in that as assets. The other that I’m most excited about is it looks like there could be an actual bar in Miami. The funny thing is this meme becoming a reality. We were joking about what would happen if this all took off. Imagine we had our own Yacht Club where we would just hang out at a bar together. And in fact, that might actually happen, and we’re going to build one. And then the next one was DOW, which has really become, I think, going to be the big, next thing in crypto. It’s been around for a while. But I think that it’s gaining steam of, okay, if you bring all these people together, how do we think about governance? How do we think about what we do with the money?
And so if you become a member of the Bored Ape Yacht Club, your voice, you can go into the discord. You can say, “This is what I want to do.” People will work together to say, “I want to do this on my own, or try to make recommendations of what actually the Yacht Club should do.” So, yes, I’m extremely excited about Roadmap 2.0. I never really thought there would be anything past Roadmap 1.0, to be honest. I thought we would do this. We would trade it. It would be this iconic thing you own that, when you looked back at time, you’d say, “Okay, when NFTs became the mainstream onboarding to crypto,” which I really believe NFTs will do, “Bored Ape Yacht Club had a really strong place in how that all happened.
But the brand and the strength has gotten so powerful, they’ve actually been able to generate something I think far beyond anyone’s wildest streams, where now we’re comparing it to Ferrari and Rolex and Supreme and saying like, “This is going to be an international, global brand. And how do we handle this and what do we do with it?” Far more than I think what we could have imagined at the onset…
…[00:20:56] Patrick: So one of the things that I’m always curious about in anything is, what is the fundamental behind the asset? And beautiful thing about Bitcoin or gold is there is none, it’s a collective story or an agreement that grows or doesn’t or whatever. And so price is kind of untethered beyond just the imaginations of donors. Whereas something like Ethereum that gets used a lot or Solana that gets used a lot or, in this case, Bored Apes, relative to, say, CryptoPunks that seem to have much less going on in their ecosystem. You can map their value onto some underlying thing that’s not just price speculation and supply and demand.
So how do you think about that as an owner? Do you care? These things are worth a lot of money today. It would be unsurprising if they were worth 5% of that amount in two months or something for some reason, because there’s not a huge amount of underlying fundamentals. So as someone interested in NFTs, how do you think about that concept of earnings equivalent or cashflow equivalent or, I don’t know what the unit of fundamental is, but how do you begin to map old models onto this new world?
[00:21:56] Eric: I think it’s really hard and it’s a question I ask myself every day looking at the prices. When CryptoPunks was released, it was released with no roadmap. That’s been a big distinction of like, we’re just putting these out to the world very much like Bitcoin. When Bored Apes showed up, they said, “We’re going to do something.” And there was a question of, what are you actually going to do? And so I think that, when I look at valuation, I would caution people that this is still wild speculation, people playing with what’s possible. But Bored Apes is by far the group that has my attention of delivering and actually building a brand and has a lot more plans for the future. And where they’ve accumulated all this money, now it becomes an execution. So without any venture capital, they’ve generated over a hundred million dollars on their balance sheet to do a what they will.
And you have a very passionate, loyal community. I think of Chris Dixon’s 1,000 True Fans just hit me. I was like, “Oh, I totally get it.” Where you’ve got 15,000 people of which there’s definitely thousands of extremely loyal fans that want to see what’s possible. And so I don’t think people are doing discounted cashflow analysis on their ape. I think they’re comparing them to other assets, if anything, and doing relative value. That’s what I’m doing. And saying, “When I first came into this space, I wanted a punk that’s our gold or our treasury we’re basing stuff off of.” And how much should a Bored Ape be in relation to a punk? How should a Bored Ape be in relation to a piece of art like Dmitri’s Ringers? And so you’re comparing it to other assets.
The interesting thing is that they actually have, like with the airdrop mechanism, they actually did start to generate cash flow, but I’m not thinking that’s going to be a professional thing. I think that’s a one time thing, because then you get to how this suddenly does start to slip for people to feel like ponzis, where you’re like, “Oh, I’ve constantly got to issue a new thing, to issue a new thing.” And I can see why people are massively skeptical, which they should be. But I think that the brand is actually going to be able to build out IP that will be valuable. And I think that we’re getting into a space that’s fascinates me because at your conference where Mauboussin was talking about intangible value. And it’s so hard for a good analyst to put that on a balance sheet. As a former portfolio manager, you’re trying to value companies, and intangibles is where all of the value’s going, but we have no good accounting metrics. And what I’m telling you is I’m buying intangible value.
I don’t know how to measure this yet. And I don’t think we have the words or the measurement at mechanisms. We’re just not at a place where you can say like, “Oh, well that’s what it does.” NFTs are starting to get into DeFi, which everything just feels like you’re going towards this asset unlock using markets for price discovery. And you’re starting to see NFTs just into the DeFi space. So with that, I think you might be able to do a little bit more of a traditional, but I don’t think we’ve figured out yet what’s the best approach to value these things. And so if anyone asks, I usually don’t talk about this stuff, and here I am on your podcast because I don’t want people to think you just buy it for $200, it goes to 150,000. Eric just DMed me the next project where that’s going to happen. We could walk away. It’s still a very speculative asset class.
[00:24:58] Patrick: What do you mean by NFTs are going into DeFi? What does that mean?
[00:25:01] Eric: So what that means is, what I’m interested is a new NFT project comes on. I’m very interested in what new aspect is it adding? We’ve seen the social status showing people, hey, I was earlier, I owned this really value asset. Clearly this is entertaining. The question is utility. What type of utility can these things bring? And so utility, to me, is innovation on the prior projects. There was a project, I don’t know who the first one to do it was, but the one that’s by far the most popular or the most successful is called CyberKongz.
And what CyberKongz did was, when you bought one of their NFTs, you started earning tokens for every period of time you held it. So if you held it for a week, you got X amount of tokens. And then after you saved up enough tokens, you could then create the next tier of the membership. And so why that’s really interesting is it’s incentivizing people to hold for longer periods of time. And then you had this economic value. You knew that if I held it for X amount of weeks, I could collect enough tokens. With those tokens, I could create and NFT and sell it.
And so then with tokens, now you can start to create liquidity pools, you can get into staking. You can do all of the things that people are doing in the DeFi space, but connected to an NFT instead of connected to a token. Now, I say that with all of the caveats I can, that what’s been really interesting about this next step into DeFi is this question of, are these securities? Are you going to have regulation? What are you stumbling into from the silly art community, this is a fun project to, “Oh wow. I have a cashflow producing asset. Was that supposed to be registered with the SEC?”
3. ConstitutionDAO, The Need for Trust, Memes and Reality – Ben Thompson
From CNBC:
Sotheby’s is auctioning off an extremely rare and historic first-edition printing of the U.S. Constitution, and crypto investors are pooling millions of dollars worth of ether to buy it. An organization known as ConstitutionDAO is raising the money using a digital crypto wallet with the aim of crowdsourcing enough funds to make the winning bid when the document hits the auction block on Thursday night.
The foundational text is valued at $15 million to $20 million. Since launching five days ago, the group has thus far raised 967 ether, or $4.3 million. The exercise offers some of the first practical insight into how crypto infrastructure can be used to facilitate fractional ownership of a physical artifact.
ConstitutionDAO is a decentralized autonomous organization, or DAO, formed for the sole purpose of putting an original copy of the Constitution back in the hands of the people…If ConstitutionDAO secures the winning bid with this consortium of crypto bidders, each contributor will fractionally own part of the text.
It’s a lovely story that has taken the crypto world by storm, and is being hailed as a demonstration of the power of Decentralized Autonomous Organizations (DAOs) to bring together people who don’t know each other for a common purpose: chip in some ether, and if ConstitutionDAO wins the auction, you own (a fractional part of) an original copy of the U.S. Constitution; moreover, you get to vote based on your share of ConstitutionDAO tokens on what you want to do with said copy.
It’s a story, however, that isn’t exactly true.
The FAQ on the ConstitutionDAO website (the definitive FAQ is this Google Doc) states the following:
Am I receiving ownership of the Constitution in exchange for my donation?
No. You are receiving a governance token, not fractionalized ownership. Governance includes the ability to advise on (for illustrative purposes) where the Constitution should be displayed, how it should be exhibited, and the mission and values of ConstitutionDAO. ConstitutionDAO is taking donations and donors are receiving governance tokens with no expectation of profit. These donations are not tax deductible at this point in time.
The most obvious reason for this limitation is regulatory: issuing ownership tokens with the expectation of a return would make ConstitutionDAO a security; moreover, a DAO is digital, and the Constitution copy is physical, and it is not well established in most jurisdictions how or if a DAO can own physical property. Then there is the fact that Sotheby’s has to follow Know Your Customer (KYC) laws: if the DAO is simply a bunch of anonymous Ethereum addresses how can the auction house satisfy its legal requirements around limiting money laundering?
Instead, according to the project’s Purchase Process Details, it appears that the actual bidder for the Constitution will be a just-formed LLC with two members/beneficiaries who may sign a letter of intent to be bound by DAO decisions, but still TBD. It’s also not clear whether or not the founders of the project will take money off of the table; again from the FAQ:
Will the core team receive any of the raised funds for themselves or get compensated in any way from this?
The core team has not received or pre-minted any tokens. Following the purchase of the Constitution, we intend to submit a proposal to be voted on by the community. While this is unusual, we believe that it establishes a precedence of mutual trust between the core team and the backers of the ConstitutionDAO.
There’s a funny word in there: trust. It sure is striking that a memetic moment meant to demonstrate the power of trustless technologies is utterly and completely dependent on trust: trust that the LLC owners won’t take the funds and run — or the Constitution copy, for that matter; trust that the money will be refunded if the bid fails; and trust that the actual DAO, when it is finally formed, doesn’t enrich the core team along the way.
This isn’t a criticism of the entire project; as I noted, for both regulatory and time reasons it’s not clear how the core team could do better, and the existence of the Google Docs I drew my information from makes it clear that the team is trying to be transparent (and said transparency has increased over the last 24 hours, including revealing who controls the wallet that holds the funds). It is interesting, though.
4. A Look Under The Hood Of the Most Successful Streaming Service On The Planet – Catie Keck
When many of us fire up our favorite streaming services, we often bump into various fury-making problems: stuff freezes, controls don’t work, or the service crashes entirely. None of these are ideal, but all seem to have become a widely understood cost of cord-cutting. For example, Disney Plus crashed its very first day because its software couldn’t handle the demand (and then it buckled again under demand for WandaVision). HBO Max is so fundamentally broken that its own leadership has admitted that the app is a mess. Even Instagram, whose Stories feature makes it a kind of streaming service in its own right, crashes so frequently it’s started alerting its users when it’s borked. Streaming can be maddening!
A service’s guts, the engineering behind the app itself, are the foundation of any streamer’s success, and Netflix has spent the last 10 years building out an expansive server network called Open Connect in order to avoid many modern streaming headaches. It’s the thing that’s allowed Netflix to serve up a far more reliable experience than its competitors and not falter when some 111 million users tuned in to Squid Game during its earliest weeks on the service…
…Open Connect is Netflix’s in-house content distribution network specifically built to deliver its TV shows and movies. Started in 2012, the program involves Netflix giving internet service providers physical appliances that allow them to localize traffic. These appliances store copies of Netflix content to create less strain on networks by eliminating the number of channels that content has to pass through to reach the user trying to play it.
Most major streaming services rely on third-party content delivery networks (CDNs) to pass along their videos, which is why Netflix’s server network is so unique. Without a system like Open Connect or a third-party CDN in place, a request for content by an ISP has to “go through a peering point and maybe transit four or five other networks until it gets to the origin, or the place that holds the content,” Will Law, chief architect of media engineering at Akamai, a major content delivery network, tells The Verge. Not only does that slow down delivery, but it’s expensive since ISPs may have to pay to access that content.
To avoid the traffic and fees, Netflix ships copies of its content to its own servers ahead of time. That also helps to prevent Netflix traffic from choking network demand during peak hours of streaming.
“We, Open Connect, bring a copy of Bridgerton at the closest point to your internet service provider — in some cases, right inside your internet service provider’s network — and that basically avoids the burden of the internet service provider having to go get it and transfer it through all these servers on the internet over to you,” Haspilaire tells The Verge.
And they’re everywhere. At present, Netflix says it has 17,000 servers spread across 158 countries, and the company tells me it plans to continue expanding its content delivery network. Netflix prioritizes where it places these servers based on where it has the most members and relationships with ISPs, the company says.
“Anyone who wants to improve performance is going to try to put a server as close to the end user as possible,” Law explains. “And in putting it there and by serving the content from that last mile network, it stops the traffic having to transit all the rest of the internet and go back to an origin. So it’s taking a load off the internet, and it’s taking a load off the peering points.”
When Open Connect originally launched a decade ago, the service started working collaboratively with ISPs on deployment. Netflix provides ISPs with the servers for free, and Netflix has an internal reliability team that works with ISP resources to maintain the servers. The benefit to ISPs, according to both Netflix and Akamai, is fewer costs to ISPs by alleviating the need for them to have to fetch copies of content themselves.
“It’s not a huge burden, but it’s certainly a relief,” Law tells me. “It’s the same principle that Akamai is founded on and the same principle that every CDN works on. Netflix’s CDN is no different to other CDNs — except their CDN is dedicated to Netflix content.”
While most major third-party CDNs do multiple jobs and manage multiple requests from many companies — Akamai, for example, says it has thousands of customers — Netflix’s internal CDN does exactly one job: it distributes Netflix content. If a content distributor doesn’t have this kind of CDN partnership or server network in place, Law says, there’s a whole lot of stuff that needs to happen along the way for you to stream a movie or TV show.
5. He wanted make his revolutionary code free. But Ali Ghodsi is now a billionaire with a $28 billion startup – Kenrick Cai
Inside a 13th-floor boardroom in downtown San Francisco, the atmosphere was tense. It was November 2015, and Databricks, a two-year-old software company started by a group of seven Berkeley researchers, was long on buzz but short on revenue.
The directors awkwardly broached subjects that had been rehashed time and again. The startup had been trying to raise funds for five months, but venture capitalists (VC) were keeping it at arm’s length, wary of its paltry sales. Seeing no other option, NEA partner Pete Sonsini, an existing investor, raised his hand to save the company with an emergency $30 million injection.
The next order of business: A new boss. Founding CEO Ion Stoica had agreed to step aside and return to his professorship at the University of California, Berkeley. The obvious move was to bring in a seasoned Silicon Valley executive, which is exactly what Databricks’ chief competitor, Snowflake, did twice on its way to a software record $33 billion IPO in September 2020. Instead, at the urging of Stoica and the other co-founders, they chose Ali Ghodsi, the co-founder who was then working as vice president of engineering.
“Some of the rest of the board was naturally like, ‘That doesn’t make any sense: Swap out one founder-professor for another?’” recalls Ben Horowitz, the company’s first VC backer and himself initially sceptical of entrusting the company to a career academic with no experience of running a business. A compromise was reached: Give Ghodsi a one-year trial run.
By Horowitz’s own admission, Ghodsi, 42, bald and clean-shaven, has become the best CEO in Andreessen Horowitz’s portfolio, which spans hundreds of companies. Databricks is already shaping up to be the firm’s best software success thanks to a recent valuation of $28 billion, 110 times larger than when Ghodsi took over. Databricks now boasts more than 5,000 customers, and Forbes estimates that it’s on track to book more than $500 million in revenue in 2021, up from about $275 million last year. It features on Forbes’s latest edition of the AI50, ranked fifth on last year’s Cloud 100 list and could soon be headed for an IPO that ranks among the most lucrative in the history of software. Already, Ghodsi’s magic act has minted at least three billionaire founders—himself, Stoica, 56, and chief technologist Matei Zaharia, 36—all of whom, by Forbes’s estimation, own stakes between 5 percent and 6 percent, worth $1.4 billion or more…
…Databricks’ cutting-edge software uses artificial intelligence (AI) to fuse costly data warehouses (structured data used for analytics) with data lakes (cheap, raw data repositories) to create what it has coined data “lakehouses” (no space between the words, in the finest geekspeak tradition). Users feed in their data and the AI makes predictions about the future. John Deere, for example, installs sensors in its farm equipment to measure things like engine temperature and hours of use. Databricks uses this raw data to predict when a tractor is likely to break down. Ecommerce companies use the software to suggest changes to their websites that boost sales. It’s used to detect malicious actors—both on stock exchanges and on social networks…
…In 2009, the 30-year-old Ghodsi came to the United States as a visiting scholar at UC Berkeley, where he got his first glimpse of Silicon Valley. Despite the collapse of the dotcom bubble nine years prior and the ongoing financial crisis, innovation was at a peak. Facebook was only five years old and not yet public. Airbnb and Uber were in their first year of existence. And a few upstart companies were just beginning to boast that their technology was able to beat humans at narrow tasks.
“It turns out that if you dust off the neural network algorithms from the ’70s, but you use way more data than ever before and modern hardware, the results start becoming superhuman,” Ghodsi says.
Unlike many foreign-born future entrepreneurs, Ghodsi was able to stay in America on a series of “extraordinary ability” visas. Once at Berkeley, he joined forces with Matei Zaharia, then a 24-year-old PhD student, on a project to build a software engine used for data processing that they dubbed Spark. The duo wanted to replicate what the big tech companies were doing with neural networks, without the complex interface.
“Our group was one of the first to look at how to make it easy to work with very large data sets for people whose main interest in life is not software engineering,” Zaharia says.
Spark turned out to be good—very good. It set a world record for speed of data sorting in 2014 and won Zaharia an award for the year’s best computer science dissertation. Eager for companies to use their tool, they released the code for free, but soon realised it wasn’t gaining any significant traction.
Over a series of meetings at cheap hole-in-the-wall Indian restaurants beginning in 2012, a core group of seven academics agreed to start Databricks. Entrepreneurial wisdom came from the Romania-born Zaharia’s thesis advisors, Scott Shenker and fellow Romanian Ion Stoica, two well-respected academics. Stoica was an exec at $300 million video streaming startup Conviva, while Shenker had been the first CEO of Nicira, a networking firm sold in 2012 to VMware for about $1.3 billion. Stoica would be CEO, and Zaharia the chief technologist. Shenker, who joined the board rather than working full-time for the company, arranged the initial meeting between Ben Horowitz, an early Nicira investor, and the researchers, who nearly balked at the idea.
“We thought to ourselves and said, ‘We don’t want to take his money because he’s not a researcher’,” Ghodsi says. “We’d wanted to get some seed funding, maybe raise a couple hundred thousand dollars and then just code away for a year and see what we could get.”
On a summer day at their new office space one block off Cal’s campus, the founders sat idly in their conference room, pondering how much money would be too much to turn down. An hour after their scheduled meeting time, Horowitz arrived. “Traffic is brutal to this Berkeley place,” he said, before cutting to the chase: “I’m not going to negotiate with you guys; I’m just going to give you an offer, so take it or leave it.” The offer: $14 million in capital at close to a $50 million valuation. It was too much to turn down.
“These kinds of ideas have a time limit on them,” Horowitz explains. “For most people, starting with seed money is the right thing to do, but not for these guys.”
6. Twitter thread on why Web 3.0 matters – Chris Dixon
Web 1 (roughly 1990-2005) was about open protocols that were decentralized and community-governed. Most of the value accrued to the edges of the network — users and builders.
Web 2 (roughly 2005-2020) was about siloed, centralized services run by corporations. Most of the value accrued to a handful of companies like Google, Apple, Amazon, and Facebook.
We are now at the beginning of the Web 3 era, which combines the decentralized, community-governed ethos of Web 1 with the advanced, modern functionality of Web 2.
Web 3 is the internet owned by the builders and users, orchestrated with tokens…
…Why does Web 3 matter?
First, let’s look at the problems with centralized platforms. (I wrote more about this back in 2018 here https://cdixon.org/2018/02/18/why-decentralization-matters/)
Centralized platforms follow a predictable life cycle. At first, they do everything they can to recruit users and 3rd-party complements like creators, developers, and businesses.
They do this to strengthen their network effect. As platforms move up the adoption S-curve, their power over users and 3rd parties steadily grows.
When they hit the top of the S-curve, their relationships with network participants change from positive-sum to zero-sum. To continue growing requires extracting data from users and competing with (former) partners.
Famous examples of this are Microsoft vs. Netscape, Google vs. Yelp, Facebook vs. Zynga, Twitter vs. its 3rd-party clients, and Epic vs Apple.
For 3rd parties, the transition from cooperation to competition feels like a bait-and-switch. Over time, the best entrepreneurs, developers, and investors have learned to not build on top of centralized platforms. This has stifled innovation.
7. Roelof Botha – Sequoia’s Crucible Moment – Patrick O’Shaughnessy and Roelof Botha
[00:25:39] Patrick: If you think about today’s landscape, setting aside the cost of capital and the crazy funding environment, which, look, is great for the world I think. There’s a lot more businesses being built and talented people building them with more access to capital. I think that outcomes will be fantastic. How else do you see today’s landscape as different than the long past in which you’ve invested predominantly here in the US? What is most notably new or different about your interaction with founders or companies or opportunity sets? What are the winds of change, so to speak, today?
[00:26:11] Roelof: I’d say one of the biggest changes I’ve noticed over the last 15 years is the scale of ambition of founders. And maybe some of that deals with people seeing examples and role models that they can follow. It’s always been powerful for me. I think most people have this experience. When you’ve seen somebody else accomplish something, it’s sort of like an existence proof from science. It shows you it can be done and it gives you a template. Part of what we lamented 15 years ago was that founders didn’t have enough ambition. And sometimes we would sell too early when they were onto something great. And I was guilty of that myself. PayPal was a 1.5 billion acquisition, and today it’s a company worth over $300 billion. And Mike Moritz was on our board. And Mike, I think, saw the potential and pushed us and challenged us to think longer term.
And what did we know? What did I know? And I made the wrong decision, and didn’t quite fathom how much potential PayPal had as a business. That has changed. And I see founders today far more patient than they ever were, far more ambitious to build companies of consequence, and also far more mission-driven than they used to be. And not only is this in America, I’ve now started to see this in Europe. I think this was especially true of European founders maybe 10, 15 years ago, where they got to a certain scale and it seemed like that was good enough. That was a great accomplishment. And now the founders really have global aspirations.
[00:27:31] Patrick: How does this most change their behavior? You’ve mentioned longer time horizons and bigger ambitions, so maybe some of that just speaks for itself. But does this manifest in different ways of doing business? Or different ways that, as a board member, you guide companies in their early years? Like you have to act differently with that scope of ambition in mind in years one through three than you would if you were going for a smaller outcome?
[00:27:53] Roelof: I think it does lead to a different outcome, because the horizon is different. You don’t take shortcuts, you don’t use duct tape, you try to architect things properly. I think teams are more mindful of who they recruit. Look, and sometimes there are people that are wonderful to help you go from year one to year three, but maybe they’re not the people to get you from year three to year five. Or they can help you get you 100,000 in revenue, but they maybe can’t get you from there to a billion in revenue. And so I see founders being a lot more deliberate about thinking about the composition of their management teams, about their willingness to invest in the next generation of products that might yield benefits down the road.
Going back to Square as an example. Square Cash didn’t exist for the first four or five years of the company’s existence. And it’s a huge business today with 10 of millions of users. And Jack has this phrase about companies having multiple founding moments. And how do you encourage the team and provide that kind of platform for creativity to blossom within your organization? That’s the kind of thing that I’ve seen change, where people are just far more ambitious, they let more ideas blossom.
[00:28:56] Patrick: I love that we’re talking during a week when Zuckerberg changed the name of a trillion dollar business and is having one of those founding moments himself. Like you said, the examples that are out there now is almost like the classic four minute mile thing. Like once someone broke the four minute mile, then all these other people started doing it. Seems like there’s something similar going on here. When you first meet, because I know you invest all the way down at seed right at the beginning of a company, and series A and beyond, when you’re first meeting with a business, with a founder, or a founding team, what are your goals in the earliest meetings? Like what are you trying to learn? What are you trying to suss out? How do you go about doing that? Because I’m sure you’ve gotten as many reps as just about anybody. So what matters to you in those early meetings with very young companies?
[00:29:37] Roelof: Difficult question, but a fun question. Every meeting is both buying and selling, just like every good interview. If I walked into that meeting feeling as though my job is to find the flaws and to trip up the founder and to be the sole person who’s judging, if I turn around and I’m actually in interested in partnering with them, they may have been turned off so much by my behavior they wouldn’t choose me. So there’s this delicate balance of how do you ask questions in an engaging way so that you’re going to answer some of your diligence questions, but you also build trust. And candidly, I’ve actually found that that in of itself is valuable. And founders come back to us and say, “You ask the toughest questions, and that’s why I want to work with you, because you sharpened my thinking. You helped me think differently about my business. And that’s the kind of thinking that I want for the future. I don’t just want somebody who’s going to be a cheerleader, patting me on the back for everything.”
So it’s being engaged. It’s being prepared. I love to understand the eureka moment. What happened? How did lightning strike? How did you think of this problem? What did you encounter in the world? Because I encounter a lot of problems. It doesn’t motivate me to start a company. Sloth kicks in. So there’s something that happened where you were so frustrated as a founder with the state of the world that it motivated you to want to go and build a company. So one of the boards I’m on is a company called Natera. It’s a public company in the bioinformatics space. They have the leading technology in the world for doing non-invasive prenatal testing. The founder and I met in high school. And he eventually came to Stanford, he did a PhD in electrical engineering. And in 2002 his sister gave birth to a baby that died within a week from a genetic condition that was undetected during her pregnancy.
And Matt went back to Stanford to learn everything he could about biology and genetics, because he thought that people needed better care. In the 21st century he found it shocking that we didn’t have better technology to help families have healthy children. That was the motivation. Yes, he was a friend of mine, but you hear him talk about the motivation for starting a company like that. It’s emotional. And you understand that this person is mission driven, and then you understand all the things that they did to uncover the nuances of why this is a problem they’re going to dedicate maybe the rest of their life to. So I love understanding that eureka moment, that insight…
…[00:35:44] Patrick: What’s the most exciting about payments looking forward, thinking back here to the early PayPal days. You’ve been in and around this part of the world for a very long time. Seems like payments is just this thriving, fascinating area of technology and technology investing. What is interesting about today’s payments landscape and what might be interesting in the future?
[00:36:02] Roelof: The beauty of payments is it helps grease the wheels of commerce. I know that’s a very overused terminology. But there’s something really valuable in that insight where it makes commerce easy. It’s one of the things that we found after the eBay acquisition of PayPal, there was tighter integration between that payments product of PayPal and eBay. It accelerated commerce on eBay. Just think about that for a second. PayPal was pretty darn successful before the acquisition, but by making it slightly better, it just totally accelerated commerce. That’s the beauty of what payments in general can do for the economy, for GDP growth. It just makes commerce easier. And so when I look around, I still see many instances where payments are broken. International wire transfers. If I want to send a birthday gift to one of my family members in South Africa. That’s an arduous process.
I love the promise of smart contracts, where you can embed a cryptocurrency with a payments event to make things far more seamless. We made an investment in Filecoin, for example. So if you think about how storage works today, you consume storage on S3, they measure how much you use, and then you get an invoice and you pay it with a credit card bill and you pay 2%, whatever, transaction fees for that. It’s kind of cumbersome. And what you get with Filecoin instead is this ability for the use of storage to lead to the payment for it automatically. It’s fully integrated. There’s no need for subsequent invoicing and disputes. It’s just fully embedded in the experience. And so I think that the promise, in my mind, of some of the DeFi technology in enabling smart contracts and lowering transaction costs, to me, is absolutely fascinating…
…[00:44:08] Patrick: The reason I chose the last one, MongoDB, is hopefully for sort of a window into your thoughts on the world of developers today. The fact that all of this technology we’re investing in and talking about so much and spending our daily lives on is ultimately built mostly by software developers and that the tools available to them have proliferated. And there’s been incredible businesses built in this space, including MongoDB. You don’t necessarily have to talk just specifically about the company. I’m even more interested in sort of your thoughts on the world of developers, how much runway there is here, whether that’s something that you’re actively interested in and your thoughts on it today.
[00:44:42] Roelof: We had a strategic insight or sort of a theme that developed over a decade ago and we internally called it the rise of the developer. And that informed our investment in Unity, which also faces developers, GitHub, MongoDB, Confluent, many of these other businesses. It’s about the rise of the developer. And the reason is that there are about 25 million people on the planet who write software for a living. 25 million people. It is a staggeringly small number of people on whom we depend to build all these wonderful products that we use today. And so whenever you can deliver technology or products that help those developers become more productive, it has a massive force multiplier effect on the world.
[00:45:24] Patrick: I think that the story with Mongo specifically, especially at second act with Atlas, this product that they’ve grown and now is, I don’t know what percent of the business it is, but it’s kind of staggering.
[00:45:33] Roelof: It’s about half.
[00:45:34] Patrick: Yeah. I mean, it’s incredible in a short period of time. What have you learned there about just the team and how they’re able to manage that? And this is kind of a question about second acts. Like how do companies successfully pull this off? You’ve seen a lot of these second acts, these second founding moments, or maybe even third. Is there anything shared and common between those key moments that you could point to?
[00:45:53] Roelof: Firstly, the focus on the developer, making the developer’s life easier. The founders, Dwight and Eliot, were at DoubleClick before they started Mongo. And they experienced some of the challenges of database scalability and complexity with traditional relational databases. And that’s what inspired them to build a product for themselves as developers. They wanted a database with a schema that was different, a NoSQL schema, a document-based, that gave them more flexibility to build applications. The founding inspiration, back to the Eureka moment comment we talked about earlier. So that’s what inspired them. They were born in an era that sort of was right at the cusp of cloud computing. And so initially they used an open source approach and it was a product that people would download and build applications locally.
And then they faced what I call a crucible moment decision. Companies, in my mind, face one to two crucible moment decisions a year. And the challenge they have is identifying that it is actually a crucible moment. Because sometimes your head’s down, focused on all the execution issues in your business. And so you need a little bit of perspective and distance to see a crucible moment. And then you have to get the decision right.
So I remember that, as a board member, we were able to bring the learnings from what we were seeing in Silicon valley and what was happening with cloud computing to make sure that moving to the cloud was a conversation that really had the focus of management. And we ended up recruiting an outside board member who was a former AWS executive who could help the company unlock getting to the cloud. Because sometimes, even if you know it’s a crucible moment, even if you make the right decision, it’s hard to change your habits.
It’s almost like when you learn a new sport and you don’t quite get it right because your body’s got a habit from a different sport. You need to retrain your muscles for this new activity. And so the company went through this challenge of trying to figure out how to become a cloud business. And how do you compensate your salespeople? What exactly are the product features and how do you continue to invest in your legacy product versus the new product and deal with all those challenges? And Dave and the team have done a spectacular job. And as I mentioned earlier, it’s about 50% of the company’s revenue today.
And again, the focus was simplicity, because for the average developer, using Mongo on prem meant that you had to download it, you had to provision a server. It just requires a lot more work than consuming a database as a service, where we take care of figuring out how to scale it for you, how to shard it for you. It helps you focus on what you’re good at as a developer and you don’t have to do what is known as undifferentiated heavy lifting.
Disclaimer: None of the information or analysis presented is intended to form the basis for any offer or recommendation. Of all the companies mentioned, we currently have a vested interest in Alphabet (parent of Google), Amazon, Apple, Meta (formerly Facebook), Microsoft, MongoDB, Netflix, PayPal, and Square. Holdings are subject to change at any time.
