From Chef to System Initiative  ## Metadata - Author: [[The Changelog: Software Development, Open Source]] - Full Title: From Chef to System Initiative - Category: #podcasts - URL: https://share.snipd.com/episode/278057f3-61fc-4776-baa5-149699ca450f ## Document Notes ## Highlights - **Bulletin Boards Sparked A Career** - Adam Jacob ran bulletin boards and FidoNet as a kid, paying long-distance fees to connect globally. - He used those systems to give friends internet access by booting into Linux remotely and sharing dial-out connections. Transcript: Adam Jacob And so she bought one that had a modem because she wanted to be able to dial into the MLS. That's the multiple listing service for people who didn't grow up the children of real estate agents. And it was where all the houses are, you know, like when they list houses, it's like the database you could go look at. And so you could dial into this bulletin board and the bulletin board was the MLS. And then, you know, I thought it was awesome. And so you couldn't keep me off the thing from both playing video games because obviously but also that modem meant you could just call bulletin boards and so I have an older brother my Older brother had a friend who he'd been on bulletin boards and so he immediately showed me that bulletin boards existed and how to call them and I was just like wait wait wait wait wait Like you can just like call other people's computers and then do stuff like talk to them you know like you could page the sysop and they would like it would like ding their computer and They would hit space bar and then you could like chat with the person who like ran the bulletin board and like i just thought that was the coolest thing that had ever happened and then i Discovered fido net so fido net was like an early version of email. So there's like PacMail. So you could build. And what was cool about FidoNet was it was basically like a regional network. So what happened is every bulletin board would call a central bulletin board, or sorry, the opposite way. So the central bulletin board with a central hub would call all the bulletin boards in an area code and then collect all their mail and then pass and then make one single long distance Call. And so everybody would sort of pitch in to cover the long distance fees for transferring your mail and you could send mail all over the world and the thing that blew my mind because this Was like peak cold war was you could send email to russia so like you could like talk with russians over fido net and so eight- old me and no one knew how old you were. And so I was like a precocious eight year old or whatever. And I thought that was super cool. You could play role playing games, like play by email role playing games, which I was all the way into. So yeah, I immediately was like, loved bulletin boards. And then was like, I got to run my own bulletin board. So I was like annoying my parents by like, because I immediately signed up for FidoNet. Right. And was like using my allowance to pay the long distance fees. And, uh, and so that's, you know, the, the bulletin board would call mine at like two in the morning and the phone and I had it. So the computer would like pick up only in that small window. And so, you know, night at 2 in the morning, the phone would ring once and wake my father up. And then he'd go back to sleep or whatever. And eventually, they were like, what the f*** is happening with this phone call happening at 2 in the morning all the time? And confronted me about it. And I was like, oh, yeah, well, oops. So yeah, I love bulletin boards. I ran bulletin boards forever until bulletin boards weren't a thing. I ran a bulletin board in high school that I gave all my friends because internet access was still rare, you know, in the early 90s. And so I would, that's how I sort of started running Linux was I'd run, I was running OS2 warp because I wanted multiple phone lines because you could multi-thread. And then I discovered Linux probably in 89 or 90. It couldn't have been 90. I don't know, early. Slackware, pretty red hat. Anyway, so I wound up like, I had a bulletin board where you could dial into it. ([Time 0:05:36](https://share.snipd.com/snip/9cc71682-0791-4cb5-9851-c00c912af752)) - **Linux Fueled Open Source Commitment** - Linux changed Adam's life by exposing source code and a generous community that taught him systems. - That open-source experience became the core reason he values licensing and community-driven growth. Transcript: Adam Jacob I mean you know when you could get it on floppy disks and like and then you suddenly had this, like, I loved operating systems. And so, but you couldn't see the source code to DOS or OS two or windows, you know, like, but Linux, I was like, Whoa, I can like, I can read like man pages for everything. And I can look at the source code and I can like figure out what to do. And I learned what a compiler was and like all of that stuff happened because Linux happened. And then there was this huge community of people who were so giving with their time and so giving with their, um, with their focus that they like, that they really were, it was very supportive As a place to learn. And so you know uh when i was working for that isp in arizona i patched the red hat installer so there's a bug in kickstart that made it so you know we basically couldn't automate the installation Of this isp that we were building and so i patched it and sent them the patch and they like accepted it you know and suddenly that was like it showed up in an errata you know and then when they Ipo'd they like sent me a little thing and we're like, hey, thanks for the patch. Like, here's your friends and family thing, you know? Like it was great. The early era of that, that era was very like, was really rich and interesting. And like, you know, it was also where I saw the first people, you know, I became a free software person. So the idea that that was important, like that's how that happened for me. ([Time 0:29:00](https://share.snipd.com/snip/d420c283-f1ed-448e-b168-f08ca3f1fe3f)) - **ISP Stack: From Modems To Red Hat Racks** - Early ISP stacks used racks of PC modems, Solaris or BSDI, Apache, Qmail, and DJB's DNS in small data centers. - Adam saved money by running Red Hat and assembling rack-mounted systems instead of buying Sun gear. Transcript: Adam Jacob Mean, what you had were sort of… From hardware to software. In the very beginning, you had in the dentist's office, it was racks of regular PC modems you could buy off the shelf plugged into like pcs that were just loaded up with serial ports right Um and then eventually there was you know specialized hardware that you would buy that just had like racks and racks of modems um and, and then, you know, it evolved over time. So in those early, in that early era, there wasn't a whole lot and it was all mostly you were running it on, on Solaris. That was, if you had money, it was Solaris. I don't know why they chose BSDI in the first, in the early days, but they did. But, you know, if you had, if you had capital, then you were running all that stuff on Sungear. Um, and so for a lot of people, it was like Solaris for all the stuff that matters. And then, uh, and then, you know, racks of modems, um, that got ever denser sort of as time went by. Um, I think peak, you know, for me, um, we were running red hat 4.2, um, in production at, at the ISP in Arizona. So like, which I, that was us. Like we were the, me and my friends were the ones who decided to do that. Um, and saved a bunch of money cause we weren't buying sun gear. And so we were just putting compute, we were putting rack mounted systems together either by hand or eventually buying them. Um, and then, uh, and then sticking them in the closet and basically, um, and then, yeah, that was like, that was pretty much of the stack. And then you were running, you know, Apache and you were running for us, we were running Q mail for, for email. Um, yeah, that was about it. You know, you got your home directory and you could like put up your little website on it or whatever. And yeah, other than that, you got access. Oh, DNS, of course. For sure. Um, so like I was a big DJB person. So there was a lot of like, you know, we were running QMail and we were running his DNS server and we were running all that stuff. ([Time 0:32:34](https://share.snipd.com/snip/087f32ae-9830-4a75-9000-cb5262521e5e)) - **Leverage Open Source To Build Business** - Use open source to create value others can adopt and then capture revenue in complementary ways. - Embrace community contributions early to accelerate product-market fit and credibility. Transcript: Adam Jacob So one was, so Miguel de Acasa and Nat Friedman. Nat Friedman would go on to be the CEO of GitHub and a bunch of other things. They started a company in the early days called Zimian. I don't think that was the original name, but basically they were the gnome company and they raised venture capital. And I remember when the news broke that they had raised venture capital and that everything that they were doing was open source. And at this point, I had already decided that the game here was figuring out how to start a business and make a bunch of money and so like you know working in arizona we were all contractors Because you couldn't get a job working for the power company right because it was all unions and they didn't want the isp arm to be unionized in that way and so we were all contractors and Uh so we built we started our own contracting company hired everyone and gave them all a 20 raise and just like took a smaller nut you know um so we were already kind of in that zone of like How do we figure out like let's figure out how to make a bunch of money basically like obviously there's a bunch of money to be had we should get some and so we so when they started Zimian, I remember being like, oh, man, it's all open source. They described how they were going to make money. And I was like, that could work. You could have all of the goodness of the openness and the sharing and that could then make you into a bigger business. And that was the first time that that entered my mind that like, what you could figure out how to do was like, build open source and then use the success of that open source to sort of catapult You up in business. And then there was a series of others that sort of moved in that early era, there was like, you know, like VA Linux, and the people that now run, that now run one of the one of the Hat clones, Rocky Linux. Like those guys had a company called Linux Care that's sort of infamous for its terrible booth babes. When you look back on it now, it's like, we're bad, you know? Wow. And like, so there was a couple of those things happening. And then Red Hat, of course, happened. And I loved Red Hat. And they treated everyone so well in that era, really. And when Red Hat IPO'd, they made a ton of people wealthy. And I was just like, yeah, that's the way. We should figure that out. Philosophical level so like because it had made such an impact on my life like i had this career i was making more money than on than any of my friends were like i was we were doing well and Like that's um that was all because of linux and it was all because of open source right that isp that i was running red hat on like they loved me because i had saved them tons of money right Um and what evolved over time was that there was like a, I love that strain of like human goodness that lives inside of open source. I love that, like that, that ethic of sharing and that like more is more, you know, like we can, we can, the pie can get bigger and then everyone can eat. And like, um, and that, it was just so, it's such an ingrained thing in me now. Um, but yeah, it was, it was very motivating and very, very real and felt achievable, but then you have to figure out how to do it. And so it was like, you know, we tried to start a half dozen businesses, you know, and, and failed. Um, eventually, um, we basically those, those same guys that I worked for for a long time at March X, um, March X went public and I had a pile of stock and I hadn't made any money from any of Really many meaningful money from any of those companies I had worked for. And so they had IPO'd and my stock basically vested. And for the first time, it was also above water. So I would have made, you know, not life-changing money, but like meaningful money. It would have been life-changing at the time because I didn't have a savings account because whatever, I was a kid and spending every dollar. But in order to sell my shares, I needed to get their signature from the chief legal officer. And there was this like coterie of executives who had been doing this together this whole time and it had all become fabulously wealthy. And so I went into their office and I was like, Hey, can you sign this piece of paper for me so I can finally sell these shares for my like decade of dutiful service or whatever? And he was like, no. And I was like, well, but I don't have any like proprietary information about it. I don't know any special secrets or whatever. Like, please, can you do this? And he was like no because it would inconvenience me and this is what he said it would inconvenience me and and our ceo and like all my friends so no and i was like just no he was like yeah just No and so i like picked up my piece of paper and i left their office and i walked out of the building and i called my best friend who had built that asp with me in Arizona. And I was like, Hey man, you hate working for IBM. I, these people obviously give about me, you know? Um, and so we should start a consulting company. I'm done. And like, I'll go find us clients and pay your bills. So you don't have to quit your job until I find us people who will pay us, but we'll go build automation and we'll build fully automated infrastructure for startups. And so I walked and started a consulting company and we built full, we started building fully automated infrastructure for startups. Um, and that became HJK solutions. So me and Nathan Hating Smith and our friend Sox Siri. And then, uh, we added some more people, Barry Steinglass and a few others. And eventually that became Chef. ([Time 0:40:10](https://share.snipd.com/snip/34478b54-f9c9-4698-aac7-26db70872648)) - **Chef Grew From Consulting To Product** - Chef began as automation for Engine Yard and internal consulting needs before a public launch and VC raise. - Ezra Zygmuntowicz supported Chef early, helping its adoption in hosting and Ruby communities. Transcript: Adam Jacob So Jesse, um, and, and if you ever meet him, he's like a larger than life human, like is an incredibly motivating person. He has a deep and foundational belief in both himself and other people. He's a lovely person. And it was Jesse who was like, you know, we tried to recruit him to our little consulting company. And we were using Puppet at the time. We built all that automation on top of Puppet. And there was this horrific moment where Puppet had two things wrong with it. So one was we were automating more than most people were with Puppet. And so the way Puppet was designed meant that it wasn't very repeatable. So you'd build this big graph, and then you would do a topological sort of the graph. And topological sorts are random, basically. And so sometimes the graph would sort in a way that worked, and sometimes it wouldn't. And the answer was just run the automation again and hope it sorts itself out. Or figure out the bug in your graph and fix the bug. Either way, what it meant was we would sell all this automation to these people and it would work 100% of the time, 80% of the time. And then 20% of the time, it just wouldn't work at all. Or it would take five times as long. And people didn't love that. And so that was frustrating. And then there was this bug where suddenly files at random would start getting overwritten by checksums. So you'd have like, you know, your resolve comp file or whatever would just go away and it would be replaced with just a checksum. Yeah, totally. That's the worst ever. It was so bad. So this happens to like our biggest client, right? And i hop on irc and i'm like yo dogs you know have you seen this horrific thing happen and the channel is like oh yeah we know about that one and like links me to the bug and i'm like what we You know like how what memo did i miss and this isn't fixed like how did this not put just how did it not pop to the top of the community stack? You know, you just, you know, like I should, I like, I was paying attention. I was pretty involved and like, I didn't know. So anyway, this bug comes in and in the bug report, there's, you know, it's filed by this kid in New Zealand and he files the bug and Luke responds. And he's like, I tried to reproduce this and I couldn't make it happen in the lab and I've spent as much time as I'm willing to spend on it so unless you're willing to pay me for a support contract I'm done um and I was like for real you know like it just overwrites files with checksums and you're just like I'm out and he was like yeah i gotta eat and i'm like yeah man i have to eat too We should fix this bug and he was like well you should pay me and i'll fix it and i was like what if we just worked together how about i put my labor um toward figuring out what the bug is and We solve this problem he's like no if you do that you're taking food out of my children's mouths and you should like, you know, you should go somewhere. Don't do that. And I was like, dude, I'm not paying you, you know, like, I like, I'm, I'm not going to pay you money to fix this bug. I'm willing to fix it, but I'm not willing to pay you to do it. I'm willing to put my own time against it. Cause I got a consulting company to run. We weren't making that much money. It's not like I had the cash laying around. And so I went to the person who filed the bug. And I was like, hey, can you get this to happen reliably? And he's like, yeah, absolutely. And it wasn't reliable for us. And so I was like, great. Can you get me access to the system that does this reliably? And he did. And so he gave me access to the system in New Zealand. And I spent a week figuring out this bug and we fixed it. And, you know, he accepted the PR, but was very angry at me because I had like undercut his model and taken food from his children's mouths. And so I was like, this probably can't stand. Like, we can't continue to be, I don't want to be the consulting company that is like the best in the world at this technology, but that also can't, but the guy who writes the technology Hates us. That's like not a very tenable business position. And they were completely uninterested in talking about the other problems I had with the system, which was like, hey, it scale, you got to figure out like how this big graph works. It's a mess. Like, and they were just they were like, you're an idiot. If you didn't if you didn't do it, if you weren't such a big dum-dum, this wouldn't be a problem. And so I started writing Chef on the side. So my partners took over all the work. And then I essentially took a couple of months and wrote chef and intending to use it for our consulting company. But then it was so cool that I showed it to Jesse Robbins and Jesse Robbins was like, we should raise venture capital for that. Like, let's go. And so Jesse joined us as the CEO and we raised venture capital. And I showed it to Ezra Zygmuntowicz at Engine Yard, who, rest in peace Ezra, who would have written Chef if I hadn't written it. He was already thinking about doing it. But I had written Chef and he was like, this is what I want. And so before we even publicly launched it, we used it to automate Engine Yard and then to make it available to Engine Yard's customers. ([Time 0:47:54](https://share.snipd.com/snip/15853ac2-0466-4e80-8169-57b4084b5c56)) - **Puppet Bug Sparked Chef's Creation** - Adam patched a Puppet bug, fixed it with a contributor in New Zealand, and felt the project's maintainer resisted unpaid collaboration. - That friction motivated him to write Chef to solve repeatability and scale problems Puppet couldn't in their use case. Transcript: Adam Jacob And there was this horrific moment where Puppet had two things wrong with it. So one was we were automating more than most people were with Puppet. And so the way Puppet was designed meant that it wasn't very repeatable. So you'd build this big graph, and then you would do a topological sort of the graph. And topological sorts are random, basically. And so sometimes the graph would sort in a way that worked, and sometimes it wouldn't. And the answer was just run the automation again and hope it sorts itself out. Or figure out the bug in your graph and fix the bug. Either way, what it meant was we would sell all this automation to these people and it would work 100% of the time, 80% of the time. And then 20% of the time, it just wouldn't work at all. Or it would take five times as long. And people didn't love that. And so that was frustrating. And then there was this bug where suddenly files at random would start getting overwritten by checksums. So you'd have like, you know, your resolve comp file or whatever would just go away and it would be replaced with just a checksum. Yeah, totally. That's the worst ever. It was so bad. So this happens to like our biggest client, right? And i hop on irc and i'm like yo dogs you know have you seen this horrific thing happen and the channel is like oh yeah we know about that one and like links me to the bug and i'm like what we You know like how what memo did i miss and this isn't fixed like how did this not put just how did it not pop to the top of the community stack? You know, you just, you know, like I should, I like, I was paying attention. I was pretty involved and like, I didn't know. So anyway, this bug comes in and in the bug report, there's, you know, it's filed by this kid in New Zealand and he files the bug and Luke responds. And he's like, I tried to reproduce this and I couldn't make it happen in the lab and I've spent as much time as I'm willing to spend on it so unless you're willing to pay me for a support contract I'm done um and I was like for real you know like it just overwrites files with checksums and you're just like I'm out and he was like yeah i gotta eat and i'm like yeah man i have to eat too We should fix this bug and he was like well you should pay me and i'll fix it and i was like what if we just worked together how about i put my labor um toward figuring out what the bug is and We solve this problem he's like no if you do that you're taking food out of my children's mouths and you should like, you know, you should go somewhere. Don't do that. And I was like, dude, I'm not paying you, you know, like, I like, I'm, I'm not going to pay you money to fix this bug. I'm willing to fix it, but I'm not willing to pay you to do it. I'm willing to put my own time against it. Cause I got a consulting company to run. We weren't making that much money. It's not like I had the cash laying around. And so I went to the person who filed the bug. And I was like, hey, can you get this to happen reliably? And he's like, yeah, absolutely. And it wasn't reliable for us. And so I was like, great. Can you get me access to the system that does this reliably? And he did. And so he gave me access to the system in New Zealand. And I spent a week figuring out this bug and we fixed it. And, you know, he accepted the PR, but was very angry at me because I had like undercut his model and taken food from his children's mouths. And so I was like, this probably can't stand. Like, we can't continue to be, I don't want to be the consulting company that is like the best in the world at this technology, but that also can't, but the guy who writes the technology Hates us. That's like not a very tenable business position. And they were completely uninterested in talking about the other problems I had with the system, which was like, hey, it scale, you got to figure out like how this big graph works. It's a mess. Like, and they were just they were like, you're an idiot. If you didn't if you didn't do it, if you weren't such a big dum-dum, this wouldn't be a problem. And so I started writing Chef on the side. So my partners took over all the work. And then I essentially took a couple of months and wrote chef and intending to use it for our consulting company. ([Time 0:48:36](https://share.snipd.com/snip/2036a7fd-730c-4a6e-86a9-94b3213f047f)) - **Open-Source Business Models Vary Widely** - Early open-source monetization experiments often fail until you align product, timing, and enterprise willingness to pay. - Hosting, on-prem features, and open-core were all tested before settling on Red Hat–style services. Transcript: Adam Jacob Yeah. So initially it was going to be a hosted SaaS. So this was also the moment, the rise of SaaS. So Salesforce had launched a few years earlier. AWS had just begun. So like S3 was just happening. EC2 had just happened. And so SaaS was just beginning. And so our idea was that we were going to run a hosted service. And then if you want to run it on-prem, which you'd be a dummy to do, you would be open source. And we were just way too early for hosted configuration management, like way too early, you know? And so, yeah, we launched this hosted SaaS product, big multi-tenant SaaS service that did okay, but didn't do great. And all of the large customers, banks and insurance companies and stuff like would just run the open source because they, uh, they, they weren't willing to run a hosted product. A decade later, they would all come to me and be like, can we just buy a managed, can I just buy a SAS? And I was like, you guys, like, like I did this for you and now I have let it founder because you didn't care. Um, but yeah, um, that was it. So that was the original business model. This was SAS. ([Time 1:09:55](https://share.snipd.com/snip/57c07f83-71b1-4b59-bd79-0fd92f19dc72)) - **Leadership's Emotional Toll Revealed** - After rallying Chef with a speech, Adam collapsed and wept alone to release built-up stress. - That moment showed the emotional toll of leading a startup and the need to process vulnerability. Transcript: Adam Jacob Yeah, maybe. And I had to give a speech. So like, cause the company was real, was really down. And so the Barry Crist asked me to like, give him a pep talk, you know? And so I went and watched Al Pacino did a football movie whose name escapes me. Um, uh, which one was it? Any given Sunday. It's any given Sunday. Adam Stacoviak Yes. Any given Sunday. Adam Jacob And the end of any given Sunday, he gives the greatest inspirational speech ever put to film. You know, he's basically got this, this bunch of football players and he's like, life is a game of inches and you got to crawl for it. You got to look at the people next to you. And, you know, he gives this like incredible speech about basically how the team needs to pull together and that, and that the struggle is living that like pushing through it is the source Of what it means to be alive. And so I like studied him delivering this speech because that's how I wanted to deliver it. And I, and I wrote my speech and I did my best to deliver it. Like I was coaching a football football team, like I was Al Pacino. And it worked. I think it did actually rally the team. It rallied the company. But as soon as I was done giving this speech, I'm so grateful that I was alone. So we had this office in San Francisco that wasn't full of very many people. And it had a beanbag because whatever startups. And I just collapsed and wept for like half an hour, like full body, just weeping. ([Time 1:18:53](https://share.snipd.com/snip/373b20b7-e122-4b78-a287-457fb1850392)) - **Protect Identity From Startup Outcomes** - Separate your identity from company outcomes to sustain long-term leadership. - Shift focus to controlling your work quality and response rather than uncontrollable results. Transcript: Adam Jacob Yeah, maybe. And I had to give a speech. So like, cause the company was real, was really down. And so the Barry Crist asked me to like, give him a pep talk, you know? And so I went and watched Al Pacino did a football movie whose name escapes me. Um, uh, which one was it? Any given Sunday. It's any given Sunday. Adam Stacoviak Yes. Any given Sunday. Adam Jacob And the end of any given Sunday, he gives the greatest inspirational speech ever put to film. You know, he's basically got this, this bunch of football players and he's like, life is a game of inches and you got to crawl for it. You got to look at the people next to you. And, you know, he gives this like incredible speech about basically how the team needs to pull together and that, and that the struggle is living that like pushing through it is the source Of what it means to be alive. And so I like studied him delivering this speech because that's how I wanted to deliver it. And I, and I wrote my speech and I did my best to deliver it. Like I was coaching a football football team, like I was Al Pacino. And it worked. I think it did actually rally the team. It rallied the company. But as soon as I was done giving this speech, I'm so grateful that I was alone. So we had this office in San Francisco that wasn't full of very many people. And it had a beanbag because whatever startups. And I just collapsed and wept for like half an hour, like full body, just weeping. Because the stress of trying to hold it all together and not knowing, you know, I just told all these people what to do, but what if they didn't, you know, what if it, what if we lost, what Would that mean? What'd that say about me? What would it say about them? Like what would happen to their lives? What happened to mine, my family and all of that pressure and all of that, all of that was so intense and I just couldn't hold it anymore. And I realized that the problem was that this had become my life and not my job. And that it was, you know, there's a part of it being your life that's helpful when you're trying to do something new, when you're, you know, you hear musicians talk about this all the Time where they didn't have a plan B, that's good and can be helpful. But eventually, it turns out that it's your job and that I don't control the outcomes. I don't control what people do. I don't control whether it's going to work. What I do control is how I act. I control what we do next. I control how we respond. And to do that well, I needed to put down the burden that said that it was my responsibility. And instead, my responsibility was to just be the best I could possibly be at the work that I had to do. And it had to become about the work I was doing, not about whether or not it was going to to happen for all of these people. ([Time 1:18:53](https://share.snipd.com/snip/3d4e0fae-02c1-4e6f-8482-c32a6fc1f673)) - **Fundamental Design Beats Surface-Level Competition** - System Initiative's edge is foundationally different architecture that is hard to copy quickly. - Competitive threats are less about single tools and more about whether users adopt a different workflow. Transcript: Adam Jacob I mean, no. If I could see them they're not a docker like the thing about docker is that nobody saw docker and then docker happened and everybody's like docker you know like it was a that was a that Was an avalanche of of like a real sea change in the experience of what was possible change and the better right would you agree for the better Yes. Okay. I would agree Docker was for the better. Okay. I mean, my spicy hot take is I don't think Kubernetes was for the better. I think we're actually net worse on most vectors, but I'm alone pretty much in my feelings on that. But that could be because I'm a grognard. You know what I mean? Like it could just be that I'm whatever, old. So yeah, if I could see, if I knew what it was, then I'd be reacting to it already. Um, but I don't know what it is. I think the biggest challenges for system initiative aren't there's some other disruptive technology that sort of eats my market share. I'm far enough ahead in terms of what the technology is and does that like, you'd be insane to try to do what I'm doing because it would also take you five years. Even having prior art, like you don't really have prior art. Like the source code is not enough to understand how it works or why. Like you have to, that's knowledge that like the people who've worked on system initiative have, but the market doesn't have that knowledge and wouldn't even from the source code. So copying, it doesn't make any sense. I think it will later, but it doesn't now. ([Time 2:01:50](https://share.snipd.com/snip/555bfde2-9e2f-4b1f-86e6-3a1928ad49f7)) - **Fix Foundations, Not Just Portals** - Don't paper over poor foundational tech with portals; fix the underlying stack instead. - Evaluate whether platform engineering masks deeper architectural problems before investing in surface tooling. Transcript: Adam Jacob Kind of settle this i kind of settle it by thinking about what people are going to do with their time. So like what system initiative is good at is like it's so much quicker and safer and faster to use system initiative to model infrastructure and to look at how the system could work or Should work. And like it's dramatic and it's much easier to extend to do interesting, complicated things. So that's going to, as a foundational technology people can use to build what they need to build to solve their problems. It's great. You know, platform engineering is a marketing response and a technology response to the same fundamental problems the system initiative looks at. You know, you look at that same problem that says DevOps teams, they're not working as well as you hope they were. They've been struggling to like deliver the results over a long period of time. And as the system got bigger. So then they have this reaction to it that says, what if in fact we were just wrong about it? And what we needed was an API all the time. We've had this conversation a bunch of times as an industry where it's like, we need portals. You know, you're going to need a, you need a developer portal, you know? And so you build a developer portal and they could click a button and they get their little development environment. And then they don't have to think about all the details and it's going to be awesome. And we do this all the time. We've always been doing it. And we've been doing it since I started working in the industry. But now they're like, hey, what if that was the answer? And so what if the answer to this fundamental experience problem that we're all having is a portal? That if we build a portal better and we make the APIs better and make the portal more flexible and we make the lines between who builds the stuff that the portal runs and who uses it better, That the outcomes will be better. That's their bet. I think it's a dumb bet because it's being built on top of the same foundational technologies that delivered the terrible user experience you didn't love. So how exactly is papering over that going to be better? It won't. It won't. Because you've already fixed the experience at the bottom. You were like, no, the shape of the bottom infects the shape of the top. ([Time 2:05:28](https://share.snipd.com/snip/280b7cfd-f664-4c12-87cf-25d203860552)) - **Programmable Automation As A Creative Platform** - System Initiative offers a programmable, shareable automation model with live validation and small functions. - That interactive loop aims to turn infrastructure into a creative, community-driven platform. Transcript: Adam Jacob Of the things that you do in System Initiative is you customize the way this big, it's got this big hypergraph of functions that is the thing that actually configures, like writes the Configuration and sort of runs a simulation of what it is you kind of want to do. That whole thing is programmable. And so the loop of how you create new assets and add new behavior is getting very tight. So like, for example, there's a, you push a button and then you can create, you know, if you had, um, let's say you needed to automate your CDN and we don't support that yet, or you have your Own application has like a deployment mechanism that's custom. You could just click the customize button in system initiative. You write some javascript that defines the properties of the asset you want and then you write functions just little short functions the longest ones are maybe 100 and 200 lines long And usually that's because there's data in them not because the they're that complicated to write that you know are what actions you would take so you want to deploy action so you would Write a little function that describes what it should do on deployment, or you have a qualification that you want to run when somebody puts some data in to tell them in real time whether Or not the configuration is good or bad or whether it would work or not. So like, for example, when you use like IAM rules in AWS for doing security stuff, AWS has a way to validate those rules. That's a, that's a qualification that we wrote for the IAM rule. So when you're, when you're like using an IAM policy, every time you change the policy, it runs the command that validates your policy and tells you if it's good or bad and shows you in Real time. It's sick. All of that is stuff you can do yourself and you don't have to ask me anything. And then there's a little button. And that little button says contribute. And if you click it, it tells you, you know, that you're agreeing to the terms of the Apache license that we're going to review your code. And then you can push the button and it'll come to us and we'll review your code and then we will ship it. And we'll do that loop over time. And so I think what's on the horizon is I think this new way of building automation gets connected to the creativity that is latent in that community where like, because the tooling has Been the tooling now for a while, it hasn't been the most creative work to like do DevOps stuff because mostly what you're doing is like, Oh, I'm writing more Terraform. How long am I writing Terraformal? Five years, you know, doing what I do. Get ops. Yeah. Doing get ops. Yeah. And suddenly there's this thing that's very different, works in a very different way and it's fun. And so I think people are going to have a lot of fun, like writing stuff in system initiative and making it do things and then sharing it with each other. And I think that that's, what's going to propel it into the, into the world. ([Time 2:12:46](https://share.snipd.com/snip/26ad9fdf-613b-4f4e-9797-b02fc294b3a0)) - **Make Customization And Contribution Frictionless** - Make automation extendable and easy to contribute to lower barriers for teams to share assets. - Provide tight interactive loops so users can customize, test, and contribute improvements quickly. Transcript: Adam Jacob So like, for example, there's a, you push a button and then you can create, you know, if you had, um, let's say you needed to automate your CDN and we don't support that yet, or you have your Own application has like a deployment mechanism that's custom. You could just click the customize button in system initiative. You write some javascript that defines the properties of the asset you want and then you write functions just little short functions the longest ones are maybe 100 and 200 lines long And usually that's because there's data in them not because the they're that complicated to write that you know are what actions you would take so you want to deploy action so you would Write a little function that describes what it should do on deployment, or you have a qualification that you want to run when somebody puts some data in to tell them in real time whether Or not the configuration is good or bad or whether it would work or not. So like, for example, when you use like IAM rules in AWS for doing security stuff, AWS has a way to validate those rules. That's a, that's a qualification that we wrote for the IAM rule. So when you're, when you're like using an IAM policy, every time you change the policy, it runs the command that validates your policy and tells you if it's good or bad and shows you in Real time. It's sick. All of that is stuff you can do yourself and you don't have to ask me anything. And then there's a little button. And that little button says contribute. ([Time 2:13:08](https://share.snipd.com/snip/1e956e54-145d-42d3-897b-4040c019404c))