Max Howell: This is for everybody who is a programmer, developer, an open source person, and people who use the internet. We're trying to robustify. We're trying to make it so it's economically viable for open source to exist. So that people like me don't have to find contract work or full-time jobs every six months in order to continue working on open source Eric Anderson: This is Contributor, a podcast telling the stories behind the best open source projects and the communities that make them. I'm Eric Anderson. Eric Anderson: Today, we meet with Max Howell, the creator of Homebrew, which is still one of the largest open source projects ever. They have thousands of contributions, both to the hundreds or more to the core project, thousands to the various packages and ancillary projects. Max created Homebrew in 2010, or maybe earlier, maintained the projects for several years. And in the conversation, we talk about how he eventually passed it on, somewhat reluctantly, to the community, who had outgrown him. And now he's looking at making Brew 2. We talk most of the time about Tea, his newest project, which is an evolution of Brew, infusing the ability for a better package manager, both on a composability front and a monetization front, using Web3 and other crypto inspiration. Eric Anderson: I'm joined today by Max Howell, who, for many of you, doesn't need any introduction. Max created Homebrew, the package manager, in 2012. It now has 13,000 GitHub stars, 300-plus contributors. It's arguably one of the most successful open source projects around. And today, it's outlived his influence. I believe I understand there's a team who maintains Homebrew, and Max is onto other projects, including what he's here to talk to us about today. He's now the CEO of Tea, Tea.xyz, and we'll get into what that is in just a moment, but I understand it's employing blockchain and the like to help developers stake their open source efforts and maybe get a return on them. Max, welcome to the show. Max Howell: Thanks for inviting me here. Eric Anderson: This is one of those exciting moments where you see one of your heroes in life and get butterflies. And you're telling me... A lot of us, myself included, one of our first interactions with software as a programmer was with Homebrew, so it's fun to have you on the show. Tell us first what Tea is so that, as an audience, we all understand what we're talking about. And then, of course, we'll go into a lot of history and how we got here. Max Howell: Sure. Although first, I just want to say that I get that hero worship myself all the time when I meet various high profile people from the software industry and open source. And I think you never expect to become one of those people yourself. And certainly, I never did. So yeah, we're just regular sorts. Max Howell: So well, Tea is Brew 2, essentially. I stopped working on Homebrew in about 2015, 2016 or so, gradually petered out altogether on it, mostly stopped working on it due to burnout. It was a huge project. Still is, obviously, an enormous project. And for many years, I was the main person working on it, not to downplay the other people who worked on it, like Mike McQuaid, who still is the lead on Homebrew. He did an enormous amount work. But for a lot of that initial period, building the community up, making it into a large project, I was the driving force behind that. Max Howell: And that was exhausting, especially because I had a full-time job otherwise in order to pay the bills most of the time. So I quit in 2015, 2016, or so. Since then, I compulsively wrote notes about what I would do to Brew to improve it because, in the first place, I made Brew because it was something I needed personally. I was working at a company, and we needed a tool that was better than the package management solutions we had, especially to try and work in a cross-platform manner. So I built it for me and for my team, and then it turned out a lot of other people needed it. So in the years since I stopped working on Brew directly, I still used Brew, and I still could see how it could be better, and I imagined ideas for what it could be. Max Howell: But people would ask me, "Are you going to make another one?" I was like, "Nah," because never in my life have I done a sequel to something I've built. Usually, I do it once, and I'm like, "That was fun," and I move on to the next thing that takes my interest. And I've worked in a lot of different industries. Well, all mostly software, mostly, but a lot of different varieties of it. In my own time, I work on loads of different things, and that's part of the reason that I've always enjoyed open source is because you can just dabble in all sorts of different things. So I didn't think I'd do it again. Max Howell: But last year, my old friend Timothy Lewis, who I met in Chicago in 2013 or so, quite old friends now. We've matured together. He's been trying to get me into crypto stuff for years. And like a lot of people, I didn't really see the appeal because I deliberately didn't go into finance. I could have. I have a lot of friends who went into finance. I wanted to build stuff. And so I didn't see that there was the possibility with what crypto was building over the last 10 years, longer. They were building something more than that. Max Howell: And so last year, I was in between jobs, and I wasn't sure what I was going to be doing. And then me and my girlfriend got pregnant, which was deliberate, and we just thought it would take a lot longer. So suddenly, I woke up one day, very soon after we started trying, and realized that I didn't have a steady job [inaudible 00:05:42]. I didn't know what I was doing with my life, and I needed something secure to help raise a family with. I was used to being able to dabble in open source, and it didn't matter if it didn't pay a lot, and then doing contracts here and there and finding startups every now and again that took my fancy and so working on that. So open source wasn't going to cut it, for sure. Max Howell: And so I was thinking along the lines of, "Okay, how can I make the open source I'm working on pay the bills?" The age old problem with open source. For 30 years, open source has been building the internet, and yet, the people who work on it famously don't often get paid. And if they do, they don't get paid well. Or if they do get paid well, it's because Google or Facebook or Microsoft basically own that piece of open source, and they have undue amount of influence on how that project is run and unfolds. Max Howell: So I went back and started looking at what could be done. I looked through all my ideas, and I came across Brew 2 in my ideas selection. Then I had a call with Tim, and he turned me on to Web3, as they were calling it now. And I saw this connection, suddenly, that evening. The package manager is this uniquely placed tool. It sits underneath all developer tools. It sits underneath the entire open source ecosystem, and we map out the entire open source ecosystem. We know exactly how all the open source projects in the graph are connected to each other, all the way down to the kernel, although we don't, obviously, compile the kernel, usually, these sort tools. But we could. Something I'm considering for the future. Max Howell: So if you understand that, you could use that graph to direct money in some manner to those people. So the obvious choice was token, some kind of token on the blockchain because then it's automated, then the fees are manageable, the fees are something you control, and you can do fancy stuff with it. And so we spent the last nine months devising how exactly we're going to do that, and we believe that we have come up with a unique way to remunerate the entire open source ecosystem, as well as creating a kick-ass package manager that is the successor to Homebrew and that is going to do new things, things that people haven't really thought of before. So yeah, that's Tea. Eric Anderson: Back up on a couple things. Tim, you were talking to Tim. Who's Tim? And maybe you can tell us about the Tea team today. Max Howell: So I met Tim in Chicago about... That was 2013, I think. And we've worked on a few projects together and tried to start a few companies together. And he went off after Chicago and got into crypto. And I went off after Chicago and went and worked at Apple for a year, which is a whole interesting story by itself, related to that famous Google tweet, which changed a few things for me. Eric Anderson: And for the audience, this is the tweet where you talked about your interview experience at Google? Max Howell: Yeah. Yeah, exactly. So feel free to ask me about that. It's a good story. Max Howell: So yeah, we always kept in contact. We worked on a few things together. And he became quite famous in the crypto community and the Web3 community. And so he's been trying to get me into this stuff for years. And like most people, I didn't see what it was. And I think people are only just starting to see that it's more than just money, that you're building on top of money, making it programmable. It's the next step in what the internet can be, when money is something that isn't controlled and can be flexible and can be programmed. You can do all sorts of interesting things with digital contracts. And immediately, most of the uses that people are doing right now is just decentralized finance. And people harp on about the decentralization as though, obviously, you should understand why you want everything to be decentralized. And unlike most people, decentralization sounds nice, but it's not be all and end all for me, not at all. But it is powerful when used correctly. Max Howell: But so, as a result, most of what people see in Web3 and crypto is just a bunch of people trying to make a load of money, a lot of scams, and a lot of, "Why do I need this?" The utility use cases are only just starting to emerge. And I think Tea is going to be an excellent example of what the utility of Web3 can be, once we go live with it. Eric Anderson: Yeah, well, maybe just a word on... The programmable money is extremely enticing, and I'm glad you can separate the value there from decentralized in general because I agree that there's pros and cons to centralization and decentralization, and programmable money is just strictly cooler than non-programmable money. So good clarity there. And then, before we get to the team, maybe you can tell us exactly how Tea works, or maybe not how it works, but maybe tell us a step further about what it is or what you imagine it to be. Max Howell: Yeah. Well I can't tell you exactly how it works yet. So we've written the white paper. So you can't be a crypto project without white paper. So initially, I was like, "Oh, do we have to be like everyone else?" I always come at everything like this. It's like, "I'll be just copying everyone else by making a white paper." But once I read a few more of them than the Bitcoin white paper, which everyone should read the Bitcoin white paper instantly because it's really quite ingenious. There was some real inventions in there. And obviously, it's super enticing how nobody knows who the people or person is who wrote the thing. Eric Anderson: Incredible, yeah. Max Howell: So there's an interesting mystery there. So I read a few more, and then I realized that, actually, by doing that much work upfront, you were making it possible for you to do a super complicated thing, like write a blockchain and a protocol. And so I was like, "Okay, let's do it." Max Howell: So we wrote the white paper, so we more or less understand how remuneration will work. Tea is not just a blockchain, and I think this is where a lot of Web3 companies go wrong is that they only have protocol. They only have blockchain. You got to remember that protocol and blockchain is just a piece of technology, and it should just be an implementation detail for your users. You have to come at these things with a user-centric mindset. You always have to, whenever you're doing a company, whenever you're doing an open source project. So are our users only going to use a blockchain? No chance. There's no way we could be successful without an excellent, kick-ass product suite that is a companion to the protocol. Max Howell: Now, our protocol is essentially a decentralized database for all open source packages. So when you release a package, you'll sign an NFT into our blockchain. There's no JPEG or anything. These aren't the scammy, JPEG NFTs. This is the real, what NFTs were meant to be, NFTs, i.e. an immutable, owned data point in the blockchain that is decentralized and cannot be undone. So it is perfect for a package release. You release package-fu version 1.2.3, stick it in the blockchain, there it is forever. And we can use decentralized storage, so the packaged [inaudible 00:12:23] and binaries can never be unobtainable. So the decentralized stuff has some tangible benefits for security and the robustness of open source in general. But also, because it's an owned NFT, we know where to send token. So I'll talk about the remuneration in more detail later, but protocol is just one part of it. Great decentralized database for all of open source. Max Howell: And we want all open source to be in there, not just stuff that Homebrew would normally do. Not stuff that packages for systems would do. I want MPM to put their packages in there one day. So we're going to build tools to make it super easy for other package managers or dependency managers to start using our protocol in the long run. But I understand this is a three to four year project. But once we've proved the validity of it for our use case, which will be the Tea package manager, I want all these other things to start using it because it's a humanitarian effort, what we're doing here. This is for everybody who is a programmer, developer, an open source person, and people who use the internet. We're trying to robustify. We're trying to make it so it's economically viable for open source to exist so that people like me don't have to find contract work or full-time jobs every six months in order to continue working on open source. Max Howell: One thing I say regularly is that I can't imagine what else I could have built, frankly, if I hadn't had to keep chasing the dollar every three to six months. And the truth is there was a lot of full-time jobs that I'd worked at, even Apple... probably would sue me... where I was doing open source while I was there. So effectively, I had two jobs the whole time. Max Howell: So anyway, the protocol is not owned by the company. The protocol is going to be released as an open source project. The IP will be donated to the world. And Tea, Incorporated is not taking any money. It would be common for it to be like an App Store. Apple take 20% of every transaction that goes through their App Store. Well, effectively, we're building an App Store for open source here. We're not taking a percentage. We're taking zero. So Tea, Incorporated has to make money other ways. And part of the way we're going to do that is build building this kick-ass product suite. Max Howell: So the product suite is the companion, kick-ass package manager. It's going to do things people haven't really considered that a package manager can do before. I'm actively working on that right now. Soon as I get off this podcast, I'm going to go straight back into it. I'm eager, actually. I'm in the middle of something important. And we're building a series of extra things around the package manager so that we can monetize the company, so that we can give things to developers that they need. Max Howell: And Tea is going to be a very developer focused package manager. Homebrew was initially a very developer focused package manager. Over time, it became user focused in general. It doesn't matter what computer user you are, if you're using a Mac or Linux nowadays, or even Windows nowadays. I was pleased to see it across the platform barriers. That was always the intention. Then Homebrew is a great tool for getting open source software that you might need. And it's not just developers who need this stuff. Over the years, I received emails from every profession, pretty much, that uses a computer. Every profession that uses a computer, there is something in that open source bucket where there isn't a good graphical app and you need something like a package manager to get it. Musicians and architects and scientists, other kinds of engineers. I got emails from all these kinds of people, and it was annoying most of the time, frankly, because I had a lot of other things to be working on rather than user support. Max Howell: But with my open source, I've always come from the perspective of patience, as much as possible. You can find great examples of me losing my temper on GitHub, and this has given me a reputation, for sure. But it's just a fame thing. Once you're reasonably famous, every example of bad behavior gets you because generally, frankly, I'm pretty patient with people because I want to understand their problem so that I can fix the problem and make sure that I don't get that support request again. It really is a selfish basis. Max Howell: I remember some of the first open source I worked on, I worked on this music player called Amarok for Linux, KDE Linux, and it was one of the best projects I've ever worked on, honestly. But I spent hours in the support channel, just figuring out what people's problems were and then hard coding solutions for that into the build system... there was always the build system that was the biggest problem... so that we wouldn't get that request again, but also because I don't want people to have stuff that's broken. Max Howell: I know that every time you get a bug report, that's only one in 10 people are going to give you that bug report. There's nine other people you're never going to meet who just gave up on whatever it was they were using. They were like, "Nah, screw it. This doesn't work. I don't want to use it." And with Homebrew, I came at it with exactly the same perspective. It's part of the reason it was successful because from the word go, I knew that I needed to help the user to help me to fix the problem. So I built in all this stuff, whenever there was a bug, whenever it crashed, I'd print a huge stack trace and tell them to open a ticket. I'd give them the link so they could open a ticket. I gave them other information I needed. And I would search GitHub just in case it had already been reported and give them any tickets I thought might help them fix the issue, so they could go there and communicate, make a community, make it so everyone felt they were part of making Homebrew better. Eric Anderson: So Max, you said something a while back that I think is important, and that was that if we provided a way for people to make money on open source, there would be a whole lot more of it. And I think that's... When people complain about the open source economy today, the economist in me is like, "Well, it's working, to an extent." If, people didn't think that there are incentives for creating open source and people follow those incentives, and some of them get what they are after, and if they don't, they move on to other things, and in a way, it works. Eric Anderson: But what you're pointing out is that there's all kinds of more software that could be written, but there's not a business model to incentivize people to write that software. In fact, when they try to, they eventually just give up because they're working for free, and they move on to other things. But you pointed out that there was all these other projects you could have pursued if you didn't have to go work on dumb things that had a business model. Max Howell: Yeah. Well, exactly. Eric Anderson: So if we can just give a business model for people to build utilities that other developers are going to benefit from, and you're saying Tea is this medium. Brew is certainly a distribution channel, and it's very normal for a distribution channel to also be a place to monetize, and so now you're offering that to the world. Max Howell: That's how we set up what we're doing, and that is our intention, and that is what we are heading towards building. So if we don't succeed, I'll be extremely upset. But that's what we're putting all our energy into making happen. And yeah, the truth is anyone who's ever released open source did it for passion, and they'd rather be working on that. And then they have to stop. These people should be building the stuff that builds the internet. If you look at the stack of open source, 90% of pretty much every Web2 company out there is open source underneath that. Every big enterprise solution is huge amounts of open source. It used to be that they'd try and avoid it. Microsoft were a huge one for trying to avoid it. But in the end they were like, "Ah, sod it." How can you avoid it nowadays? And there's no one that isn't. Max Howell: So these people build the stuff that powers the internet, and yet, they have to either beg a company to hire them so they can work on it... And we want that to stop. It used to be open source was always maintained by people who had no external pressures that were forcing them to do things that maybe didn't make sense for the project or the community. I don't like that Facebook, Google, and Microsoft hire so many open source people now. But so we want to make it so that these talented engineers at Microsoft, Facebook, and Google aren't working on making ad algorithms or encouraging people to click a like button, but they're working on things that genuinely benefit humanity, the world, and the internet. These are talented people that need a system that incentivizes them to use their talents for what's best for humanity. And that's part of what we're trying to make Tea able to do. And honestly, I think we've cracked it. People should go read the white paper. Eric Anderson: So let's run that to ground. To get involved, now you understand, at a high level, what Tea's trying to do, and then is there a place to contribute, get involved? What's the call to action there? Max Howell: This is the first time I've made an open source company. It has been difficult. It's been a lot easier in the last few years. The idea of commercial open source has really started to become a thing, but I couldn't successfully do it before. I remember some VC phoned me after Homebrew had been out a few years and it was getting up to tens of millions of users at that point. It was like, "How can we monetize this?" And I was like, "It's too late, mate. You should have called me four years ago." Eric Anderson: Ship's sailed. Max Howell: What can we do now? So we done it in the right order. But as a result, usually, I wouldn't have had any discussion about what I was working on yet. Whenever I do open source, I do it without telling anyone, maybe a few hints on my Twitter here and there. And then when it's ready, I release it. So it's not ready, and I haven't released it, so people can't really get involved yet. We have a Discord, and I'm not great on the Discord yet, and that's probably because we haven't really got anything out there. But our white paper's open source, and you should go and check it out because I've built this hugely awesome build pipeline for turning markdown into what looks a normal PDF, using loads of open source, obviously. And our website is open source because I want everything we do to be open source, every single part of it. So you can check those out, but you can't really contribute yet. Eric Anderson: Good. Now, you also said something about these open source folks have to go get a job, and often, those jobs are at big tech, and they get paid for doing dumb things. Tell us about your experience interviewing at Google. And I'm curious to actually hear the fallout, how that tweet... You said it changed your life. Max Howell: Yeah, well, there's no doubt. It was a pivotal moment. Max Howell: So okay. I was in Chicago, and I didn't know what to do with myself next. I'd recently handed Homebrew over to the community by transferring the repo from my username to the Homebrew organization, which took me about 12 hours of painful deliberation to do, I have to say. I remember I had the page open, and there was a button on the page that was like "Transfer." And there was like, "Are you sure?" And then a button, "Okay." And I showed it to one of my coworkers at the time. I was working at a bootcamp, and I said, "Should I do it?" And he said, "Max, there is no way you should do this. Are you insane?" And then five minutes after that, I did it because I realized that I was only not doing it for my own ego, and I enjoyed having it on my username. At the time, Homebrew was the biggest open source project on GitHub by every metric. And it made me feel good. But I knew that my effort had been reducing over time, my motivation was waning, and I needed to hand it over to the community. There was loads of people who turned up who were doing a really great job, better than I was doing at that point. Max Howell: So I'd done that, and I was like, "What am I going to do with my life now?" So I asked my wife at the time, and she said, "Why don't you reply to one of those emails that Google always sends you?" because they'd been sending me recruiter emails since Homebrew had become popular, successful, and well used. Yeah. And I'd always ignored them because I had never thought I wanted to work somewhere like Google. And I was right, turns out. Max Howell: So I replied, and the recruiter phoned me, and he said, "Wow, okay, based on your status, we're going to skip the phone interview and send you straight in for in-person interviews." I said, "Well, okay." And then I asked, I said, "Look, you know I haven't got a computer science degree, right?" And he said, "Oh, yeah, yeah, that won't be a problem. Your interviews will be catered to you." And so he lied. Because I knew the Google interviews were heavy computer science, but I don't have computer science. I have a chemistry degree. I was going to be a scientist. I was always going to be a scientist. And I did one year of chemistry and discovered that it was really boring and I didn't want to do it. So I found open source, and the rest is history. Max Howell: So I went in for an interview, and the first interview I had was the binary tree question. And I don't think it was actually a binary tree question, with hindsight. I figured it out, and then I went home, and I tried to figure out how to do it, and I figured it out, using Google, obviously. I wanted to prove to myself I could do the thing. It was array-based, sort of binary tree because you had two nodes for everything. But yeah, I don't really remember at this point. Max Howell: Yeah. So I had that. And then at the end of the interview, I said, "So you don't actually use this stuff, right, at Google?" I don't know why I said this. And she said, "I use it all the time." And her attitude was one of, "I'm not going to let you get this job," as far as I can tell. I don't think it was just her. I had seven interviews that day and a lunch interview. They don't half lay them on. And I think half of them thought that I would be a good fit, and the other half didn't, and then in the end they decided no. Max Howell: But I don't want to... Google don't deserve the shit they got for it. It's one of those typical viral tweets. They phoned me up the next week and told me I didn't get it, and so I just immediately went to Twitter, and I was like, "Blah, blah, blah," because I was a little bit annoyed about it, even though I, at the back of my mind, never thought that I was going to take the job anyway, wasn't sure. And at the time, I only had a thousand followers on Twitter, so I didn't expect anything to happen. But 20 minutes later, my friend calls me, and it's like, "Where do you think your tweet is on Hacker News?" Max Howell: I'm like, "Oh God." So I went to Hacker News, and then right at the top is, "Max Howell didn't get hired at Google, told them to fuck off." I was like, "Jesus." I was like, "Should I delete this?" I almost deleted it. And then by the end of the day, I would open up the Twitter app, and I could scroll, and the whole screen would fill with new reactions to the tweet, and then I'd scroll again, and it was immediately new reactions. It was going insane. At this point, I think it's had 3 million views or something over the years. It still gets retweeted as though it happened yesterday. This was 2015, so quite some time ago. Eric Anderson: Amazing. Max Howell: So the reaction from the internet was predictable. I should be able to invert a binary tree. What the hell am I doing? Or, "No, he's proved himself. He doesn't need to invert a binary tree." But my attitude still is I don't need much computer science, and I've managed to make things that are significant. When I do need computer science, I look it up on Google, and I implement it. I can. Just because I can't do these puzzles spontaneously, for me, doesn't make any difference. And so whenever I hire staff, I don't give them those kinds of puzzles. And Google supposedly changed how they do interviews because of this tweet. Supposedly, there was a town hall, and they talked about it. And to Google's credit, another department then offered me a job. But I also had 230 other offers of jobs in my inbox. So that was why it was life changing. And one of them was Apple. And so I went to work at Apple because I hero worshiped Apple. Eric Anderson: Wow. What a way to interview for Apple at Google, basically. Max Howell: Turns out all you have to do is work there for a year, and after that, you no longer want anything to do with them. I thought they were stupid, like everyone else did during the '90s, and it was only when they released the Intel Mac Mini I became interested. I was like, "Okay, so this thing's proper now. It runs a proper processor." And I fell in love with OSX for many reasons. So working at Apple seemed like a dream come true, but I worked there a year, and it didn't go well. No. Eric Anderson: Okay. And now you found your true home, maybe, back doing Brew 2. Max Howell: Yeah. Well, the truth is I'm too entrepreneurial. I like doing my own stuff, and I like don't slowdown. Big companies have slowdown, and that was the worst thing about Apple for me is that they wouldn't let me work the way I work. They wouldn't let me be efficient. There was too much process. And I get it. why should I expect to be treated differently? But well, they did imply I would be. Eric Anderson: Huh. Well, and how do you work, Max? I mean, what does that mean, you couldn't work the way you work? Max Howell: Over the years. I've just developed a lot of strategies for being efficient. As far as I'm concerned, I give myself a task to finish by the end of the day or the end of midweek or the end of the week, and I realize when I'm not making enough progress, and I know how to cut. I know how to reevaluate my approach. I know when to give up on stuff and throw it away. And as a result, I always meet my targets. Max Howell: At Apple, it was impossible for me to work that way. They were like, "You can't do that. We need to review everything every step of the way," or very reasonable requests. But I found it very depressing being forced into that bucket, having never before been pushed into that bucket. Now, I get you work at a company, you can't be that way. So I don't really blame them. And chances are it would've been similar at Google. But for me, 10x engineering is about efficiency. Well, I don't really consider myself a 10x engineer because I don't really like the term, but the truth is I get things done a lot faster than most people, and it's just that I know how to work. And so they wouldn't let me. Eric Anderson: No, this idea of never missing deadlines, instead of pushing deadlines to fit the scope of the plan, you adjust the scope to fit the deadlines as you work. Max Howell: Well, I don't- Eric Anderson: So you're always shipping on time. Max Howell: I don't make my staff do that, although I do encourage results, results-driven development, essentially. At Tea, at least in the engineering department and the product teams, there's no set working hours. But I want a global team anyway, so that makes sense. I want someone in every time zone. But I don't care when you work or how much you work, as long as you deliver on time, having just said that I don't enforce it. I don't get upset if they don't deliver on time, but there's a high expectation that you try. And I tutor them on how to go about it using my strategy, but I know that my strategy works for me. It's not going to work for everybody. You have to figure out your own. You have to figure out how it works. Eric Anderson: Sure. But I think it's reasonable to have company values that, generally, everyone agrees on and aspires to, and on-time delivery seems like a reasonable one. Let's go back to the team. So we tabled that for a moment. You reached out to Tim, and you came up with some of this together, or at least he was in the early days. And then others joined? Max Howell: Yeah, I called him up. I was like, "I want to build Brew 2 with a blockchain component so that, somehow, we figure out how to redistribute token to the top of the stack and all the dependencies." And this was the key because, nowadays especially, you can do sponsorship and bounties for an open source project if you want. Now, I don't particularly like them, and so I've never really done them myself because I feel that you have to chase the sponsorship and bounties, and then you're introducing some agenda into how the project works. But either way, the fact of the matter is sponsorship and bounties only reward the top of the stack, the favorites, and there's tens of thousands of open source projects. Max Howell: Log4j was a great example of this in December. You remember the incident? There was a zero day in this Java library called Log4j, and it turned out a shit ton of enterprise solutions used this library and were instantly rootable, and no one really ever heard of it. Typical Nebraska problem. That's what we call it. The Nebraska problem is based on the XKCD comic, and it's the dependency tree, the tower. Right at the bottom is that tiny little column holding the whole internet up, and it's maintained by this guy from Nebraska, thanklessly. And Log4j was a Nebraska project. So suddenly, it had all this attention, all of it negative, all of it abusive, frankly. Not everyone's abusive on the internet, but most people, if they feel anonymous enough, will be, is the truth of it, especially on platforms that can't control it. Max Howell: And yeah, they don't get any funding. And I wonder if they still do, now that they've fixed it and everyone's just moved on. So anyway, the idea was use blockchain, use open source dependency tree. Because the package manager knows the whole tree and people are incentivized to maintain that correctly, we can distribute the token all the way down. So I phoned him up. I was like, "Yeah, okay. This is the idea. Came to me in a moment of inspiration last night. Had a little weed, I have to confess. And it just popped into my head." Max Howell: The moment it popped into my head, I woke up my girlfriend, who was three weeks pregnant at that point, and I was like, "I just had what I think is a brilliant idea." And she was like, "What is it?" Well, I told her, and then the next day, she couldn't remember this at all, the first person I told about it. So I told him, and he was like, "Oh, okay. I'll get back to you." I was like, "Is that it?" Max Howell: A week later, he was flying me out to meet some people, and that's when we formed the company. So we're about 13 or 14 people at this point, and we're still hiring. I need more developers. I need people who can help me with the package manager. Specifically, I very much need someone who can help me with Windows for the package manager. But also, we're building out. I keep saying it's going to be different. It is different. I don't really want to spoil it. I think it's doing some really exciting and transformative things. Max Howell: Essentially, I'm trying to make the idea of package manager even more of an implementation detail and making packaging essentially programmable so that people can write... essentially making packaging into a set of primitives, so that you can build entirely new things with a set of well-defined primitives for packaging that can install all the base... You can write scripts of this thing where it's like, "Okay, I want to use Node, and I want to use OpenSSL, and I want to use Rust, and I want GPG." The script will just install all the shit for you in a sandbox, and then you can program with that stuff. Everything has well defined APIs. Anyway, so I'm spoiling it. Eric Anderson: So it's composable, to a degree. I mean, I've come across Nix some, which is tricky to use. Nix maybe has some of that flavor, I wonder. Max Howell: Nick's should be way more successful than it is because it's doing some of these things. They fail at the user experience. User experience is beyond bad, worse than Docker. Worse than Docker, as far as I'm concerned, is... So I can't even be bothered to use the thing, even though it's powerful. So sorry, Nix. I'm planning to replace you. Eric Anderson: Good. And you've got this bright team to do it. Got it. 13 people. Not quite in every time zone yet, but that's the goal. Max Howell: We're still too America-centric, and that's just what happens. I live in America now. I've lived here for 10 years. North Carolina at the moment. Been all over. Very much enjoying the triangle area in North Carolina. If you live nearby, please give me a call. I want to meet interesting people. Been hear about a year. So it is typical. Your network is the people you've met, and typically, that's in your country. But I want someone in every time zone, so we're actively looking outside of North America. But yeah, so hopefully 20 by the end of the year. Need more devs. And so feel free to check our website. Eric Anderson: Does that have some practical purpose, like somebody's on support or somebody's available at any hour, the company's always running? Or is it more some other ideal, like it's just cool to cover the globe? Max Howell: There's several reasons. Primarily, let's be honest, I think it's cool. Secondarily, open source is global. Open source is not America-centric. So with Tea, I'm planning to release it with translations to as many language as I can have... it's very rare that command line tools do this... and translate everything. And we got a translation process in place for the white paper, and it's semantically versioned. It's lovely. Sorry. I gush about how we're doing this. Max Howell: So we've had a lot of interest from India and Russia, and I love that. I want people to be getting actively involved from other cultures. Open source is meant to help everybody and not just the Americas. And it's become more and more American-centric since I've been in it. When I started in open source, KDE, this desktop environment for Linux, if you've never heard of it, was mostly based in Europe. Nowadays, there's not much open source which is mostly based in Europe or Asia or Australia. Things have gone wrong. I think it's the- Eric Anderson: Well, it follows the money. It follows incentives. Max Howell: It's the way it's become more corporate. And so we're deliberately, with Tea, going to try and push it out of that. Like I say, I want people at Facebook to quit and work on open source full time. That's that's the golden model. Eric Anderson: Imagine the future. Max, this has been a fascinating conversation. What have we not covered that we wanted to get to today, or that you wanted to get to? Max Howell: I think we covered everything I want to get to. We didn't discuss how the remuneration works, but that's fine. Go and read the white paper. It's a staking model, and if you don't know how proof of stake works, it makes it a little more difficult to make it sound like it's going to work. But proof of stake allows you to reward the people in the ecosystem without doing energy-intensive proof of work calculations. And so some of those people in the ecosystem we're rewarding is the open source maintainers, the people who own these NFTs that they put in the tree. That's the basics of it. But the white paper goes into a lot of detail. It's got a pretty good abstract. I'd go and read that. Eric Anderson: Awesome. Max, pleasure to have you on the show. There's so many good things to cover in this, we're going to have a hard time writing a summary and bubbling it up for everyone, but we're going to do our best. Max Howell: Great. Well, thanks for having me. I had a lovely time. Eric Anderson: Have a good day. Eric Anderson: You can subscribe to the podcast and check out our community Slack and newsletter at contributor.fyi. If you like the show, please leave a rating and review on Apple Podcasts, Spotify, or wherever you get your podcasts. Until next time, I'm Eric Anderson, and this has been Contributor.