andy lapteff 0:16

This is the Art of Network Engineering. Where technology meets the human side of IT. Whether you're scaling networks, solving problems, or shaping your career, we've got the insights, stories, and tips to keep you ahead in the ever-evolving world of networking. Welcome to the Art of Network Engineering Podcast. My name is Andy Laptev, and I am once again coming to you from Munich, Germany, with the man, the myth, the legend, one of my favorite people, and I mean that. William Wilhelm. Will I? How you do it, William? I'm doing good. You're one of my favorite people too. How's it going? We just had like a half an hour show before the show. Yeah. I'm complaining that I'm behind on time and then we're just gabbing. Well, I mean, one thing, you look well rested. So if you're thinking about starting a podcast and if you're going to travel with that podcast and do remotes. And if you're a pyromaniac. Yeah, I started a fire. Um, don't land morning of and then record six hours of episodes like that day. Maybe come a day early, maybe get some rest, maybe have a little workout and a nice meal, and then space them out normally. But this is how I roll. So, dude, it's great to see you. I think we have a very fun week planned at Autocon 5. I'm going to hopefully finally look at a Python script and understand a little bit of it. That's just my goal in life, like technically.

unknown 1:38

Yeah.

andy lapteff 1:38

I just hello world. There you go. I just want to look and like, right? I can read English, like I can read a book, right? Python's supposed to be easy-ish. So I got two days of stuff I'm going to sit through and then I'll be at my booth um doing the thing. What you and I are going to talk about, sir. Lay it on, lay it on heavy. I asked William a question about code repositories and GitHub. So after a lifetime of not being able to code and crying about it, ad nauseum, and then I finally embraced it, and that helped my career tremendously. Uh, in the past six months or so, I have been creating software with cloud code. And I'm trying to do the right things, like some spec-driven stuff. There's the security thing that now looks for vulnerabilities and like, so I'm, you know, I'm I'm trying to, and I get it. Six months ago, I was yelling at vibe coders with Erica Dietrich, like, you don't understand code, you have to understand it. What are you doing? And six months later, I'm like, oh my God, look at all this stuff I'm making. I don't understand how it's built. But I think the tooling has got so good that it works. Now, is it slop? I don't know. Is it full of vulnerabilities? Maybe, but that's not what we're talking about here. But that's been my journey. And now I'm putting my code in GitHub. And I've, I think, created five or six, I don't even know if they're apps, but five or six pieces of software that do things. And these have just been things in my everyday life. Like my nephew runs a business and he had a problem, and we started talking about it. And two hours later, I had software built for him that I'm about to migrate to a cloud because it's on my home server right now. And they're gonna use it and they're probably gonna make a hell of a lot more money in what they're doing because it's just a better solution than what they pay 15 grand a year for for these. Long and short of it is I have five or six GitHub repos, and they're all private. Now, you and I have had conversations in the past about like open source, and I I forget what episode it was. I meant to pull it up and pull a quote or whatever, but I know that I was like, why would you ever put your ideas out there for people to take? And and you had really good, I guess, pushback. So where I'm at is I have these private repos, and I asked you, should I make them public? Because I am, I don't want to say I am coding. I don't know what to say. I don't want to say I am vibing, but I am creating software that does cool stuff that doesn't exist, and the code is sitting in private GitHub repos. Now, when I looked at my GitHub, what what's the little what's the little green dot thing? It shows like what you're working on, like it shows you're contributing or basically if you look at my my GitHub, it doesn't look like I'm doing anything because it's all private. So I'm like, oh, I should make these public. And then I'm immediately fighting with myself, like, but wait a minute. These are my ideas. I spent a ton of sleepless nights building this. It's something I could charge people for. Why would I release those publicly and let people take my ideas? So I don't understand why software are manifestations of ideas, right? Somebody has an idea or they have a problem they solve, and here's the thing. Why would I make my repo public and give those ideas freely away? Does that make sense?

William Collins 4:37

It makes perfect sense. Absolutely. It's crazy to me. Yeah. I could go back in time almost. I think that's a good idea. I want to go back in time a little bit. And by the way, Red Bull gives you wings. Um marketing. We'll talk about that later. So basically, when you think about it, I've had a blog for like a long time. And at some point, I think it was 2016s, 2017s, maybe, I decided to transition that blog and manage it completely as code. So Hugo using, you know, a theme, and I had a lot of custom CSS, custom short codes that I was building at the time. I already don't know what he's talking about. But basically the formatting. Okay, I'm gonna make it. Hugo boss, the genes? Hey, it might as well be. Is that if you can get those suckers on core?

andy lapteff 5:17

I'm not gonna annoy you with all the questions, but like is Hugo like a language?

William Collins 5:19

Yeah, it's built on Go. It's a basically to statically generate. It's kind of how you you don't use Hugo, I don't believe, on your new site, but the same kind of principle statically generated. So when when I built this thing and I moved it over, man, I was I probably put more time into that than anything I've ever built. It was because I'm not a front-end person, and I at the time I was figuring out like what is good for front end, what is actually practical, and what do I actually like to do? Which is the UI, right? Yeah, the UI.

andy lapteff 5:44

What were you on before? WordPress?

William Collins 5:45

Where was your blog? WordPress.

andy lapteff 5:46

Yeah, that's where we all were. That's where our site was until three days ago.

William Collins 5:50

Yeah, moving onwards and upwards. And so what I decided to do is I made two two Git repos, Andy. I had one that was essentially a private repo and one that was a public repo. Now, why do you need two repos for this blog that I just built? Well, I was all like, oh, I put in a lot of work to make this nice and polish it up. I don't want a whole bunch of people to actually clone the back end, like whatever makes it work, and then use it as their own, because then there's gonna be a lot of folks out there that have a blog that looks exactly like mine. I want mine to be a little different, which isn't a problem today because since they have AI to use to build stuff, they're gonna want to differentiate themselves and not copy, you know, at least in this scenario, at least I would have to do that. Well, I'm with you so far. You spent a lot of time making it look a certain way, and you want that to be a differentiator, making sense. Exactly. And then at that point, too, I was working out the release. So all the stuff was private, and I had a GitHub actions, I had a pipeline that would build, stuff, you know, go through all the checks, it would check for dead links and all my code, it would do these different things, and then it would publish the final thing to that public repo, like the actual site in and of itself. I'm I'm like drastically oversimplifying this, but that is what is public. And basically, with the public artifact that you get out of all this, there's a few things that you can reverse engineer from that, like a few obvious things. But when it goes back to the customizations that I built myself, the short codes, the embeddings, and some of the you know trickier CSS things, all that stayed private in my repo, and you could not extrapolate that from the public artifact.

andy lapteff 7:27

How would you even know to separate them? Because you know what you know, and I don't like I wouldn't know what code is my back end.

William Collins 7:34

Like that's where I'm at, right? Like you've landed in telco when you started tech, right? Yeah, so it's all about experiences and what you have access to work on. And I consider myself like very fortunate when I was like earlier on in my career, I happened to work with a lot of folks, and my technical roles were moved to being adjacent to network engineering as well. So I was getting a lot of practical experience at big companies with security teams, with development teams, even before DevOps existed. I was working, I forget what we called them back then. We were doing all of our application delivery on F5. So we had like a DB tier, a web tier, an app tier. And then we had these pockets of developer teams that were in charge of the applications, and they would have to go through and scope with, you know, my team, like, okay, what does the VIP need to look like? Okay, this client, they're not okay with SSL being terminated on the the middle F5. It's got to go all the way to the server, can't be terminated on the other.

andy lapteff 8:31

To be clear, you were interested, right? Yeah, because when I worked at FiServe, I was surrounded by brilliant people like this, and I was like, no, leave me alone. Like, you people are aliens. I just want to go put a new prefix list on my route map.

William Collins 8:41

Like, you know, I'm the on it, I'm the opposite. Like, I almost want to work on things I don't understand more than things I do understand, which is a it's I it sounds like a good thing right now, just like a second, but it's not a good thing. In practice, it's awful. Because you really have to I still believe that you have to pick a skill in this industry, like one of the foundational skills, and you have to actually learn it, whether that's network engineering, which means that you ought to know a thing or two about TCP IP and different things, but initially, right?

andy lapteff 9:09

Yeah, because it seems like we're heading more toward generalists, yeah, like now, right? Yep. I don't think you can start as like knowing a little bit of everything, right? Like I don't know if that's a job. Maybe help desk? I don't know.

William Collins 9:20

Yeah, but it's help desk. I mean, you're like not even an inch deep and like 50 miles wide.

andy lapteff 9:25

So you separated your back end from your front end and you hid your back end in a priority proposal. Makes sense, and I'm with you. Was that the end of the story? Because I have a connector to that, but I don't want to cut you off.

William Collins 9:36

Well, what I was gonna say is my thoughts on this drastically changed when I so like at that point, like I was working on a lot of things with Git and software development, but not in a big scale capacity. And I had never worked with anybody at that point that was a true open sourcer, if that makes sense. And understood the principles and the practices, I understood how to use Git, and a lot of these things were becoming more popular in mainstream. But once I started surrounding myself, or I didn't surround myself, I just happened to land with folks that had different viewpoints on this, so I was able to take their viewpoints of I I remember having a conversation with somebody at one point. I'm not gonna say who it is because I think a lot of people would actually know him. But he he asked me a question point blank like, why would you ever keep your repo private unless you did you unless you had to, unless there was a compelling reason? Money.

andy lapteff 10:25

Well that time you can think about that, but like we were gonna pretend and have like a fake WWE standoff, like right, just to make it more interesting, but or too authentic to like play that nonsense TikTok game. But if time is our greatest asset and most unrenewable resource, and in our careers we give time for money, yeah, that's why I would keep it private because I can monetize those ideas that I worked on with my time. So the open source people, I don't get. I appreciate that Linux is open source and my server at home I'm developing on is like I get it, and that's great. And maybe Linux Torvald is loaded in other ways, but I can't see how me open sourcing all my I'm sorry I interrupted you, but open source bothers me because I think what you're telling me to do, and this is being slightly tongue-in-cheek, but you want me to open up my private repos and give my ideas away. Now, I don't know why I would do that because they're my ideas and I can charge for them. But I also don't want to be a professional software developer. Like so I'm I'm kind of in this weird, I don't know what to do. Like the website made sense because it's public, and what are you gonna do? And like it's there anyway, and like sure, like have at it. And I like now that you can open an issue or do a PR, like a pull request, and then fix stuff. And like if there's a dead link or something I missed, yeah, somebody else can do it. I'm a one-person shop. So if the community, the thousands of people who are involved and see a problem want to report it, it's a reporting structure, it's a thing, right? Like, I just interviewed a service provider guy right before you, and I'm like, oh, you know what? The resources section I built on the site has no service provider. I need to add that. But he could go in and submit some stuff, and I'm like, oh yeah, like approve, and then the magic happens. Why the hell would I make my ideas public?

William Collins 12:12

I think that that's the wrong question to ask. And the reason is going back to like we all, everybody in this building, we all I mean, a lot of us have families, kids, we're paying for a bunch of stuff. Times are hard right now. It's not exactly a you know flourishing, beautiful green pastures. But what I would say is so I actually just came from a startup that just got acquired lumens. Congratulations, Alkira. Awesome, good times. William's now my rich friend. Yeah. I would so basically I learned a lot kind of leading up to working there. And then when I worked there, and one of the things that I think I came away with, even though we didn't have uh necessarily an open source product, but just being in some of these circles and getting talk to talk to like more business people, like on the VC side. You know what the like the definition of a moat, you you hear this a lot in those circles. Everybody talks moats. Yeah, moats is like the big thing, but I I hear a lot of people get this wrong all the time.

andy lapteff 13:05

We don't have castles. Everybody's building freaking moats now. You should see the back of my guard because I got a castle. Dumb.

William Collins 13:11

The moat thing bothers me. Well, here's where here's where everybody gets it wrong. Because I I hear this like it just annoys me. But a lot of people basically take the idea of a moat and they map that as like a one-to-one mapping to like their code or their product or like a new feature or something that they're building. So pause.

andy lapteff 13:31

For the young kids who weren't around when castles and dragons and shit were around, a moat was the big body of water you put around their castle. Yeah, you dug a tunnel around the castle to create a physical barrier so people just couldn't come in, right?

William Collins 13:42

Exactly.

andy lapteff 13:42

So why are all the software developers using moat as a metaphor now?

William Collins 13:47

I believe that software developers really don't care, usually. But the higher up, and this is where it happens. So say Silicon Valley, and this is why you see like a lot of companies get an influx of capital to execute on an idea quicker, if that makes sense, because you want to get it out to market. Timing is so important here. First mover advantage type thing.

andy lapteff 14:05

Yeah. Speaking of first to market, yeah. I am building a platform that's going to take me an enormous amount of time and effort to create. It's private because by the time I release it publicly, I need to be far enough down the road that people can't just fork it and catch me. So I agree with you, right? Like there is a first mover advantage. Why the hell would I make that repo public? Because your body said, Well, why would anything not be public?

William Collins 14:33

Like this is the there are reasons not to do it, by the way. There are reasons, but I think the the reasons in the way people frame it is just a little bit backwards sometimes. But kind of what I was getting at, like HashiCorp would be a good example here. So Terraform was open source, and you could say, okay, like Terraform, the code is so good, and they built a moat around the code that was in Terraform, then why didn't anybody fork it at that time and just start doing stuff? Well, it wasn't just the you know, the product and the code. What it was oh, what what do you call the network effect? So they built this, but the people, the practices and the execution surrounding the technology are really where your moat comes from. So in HashiCorp's example, what they did was they rallied the community, they got a whole bunch of people using it. So the more traction they got with the people, people started building modules. They had a module registry to make using the tool easier. People develop for them. Terraform registry, free marketing, free software engineering. And then beyond that, vendors jumped in. So vendors started building more providers because they saw that okay, the the users are using this. It's obviously important. Like, oh, they're using this in enterprises now. We need to get behind it. We need to have something here. And so when you take into account all of these things, that's your moat. You really, what a moat is, if you wanted to define it, it's really the amount of separation you have between your idea and your execution of your idea and the incumbents, you know, the people that are who are trying to copy you. But a lot of times you can't. So nobody forked Terraform and made a better well, it is has been forked because of light, you know, the whole open tofu movement and everything. But I would say that a lot of companies today still use Terraform proper.

andy lapteff 16:15

Is this a long game where you hope that so many people adopt it that you eventually charge for support?

unknown 16:21

Right?

andy lapteff 16:21

Like that's what the Ansible thing is.

William Collins 16:23

It depends on what it is, too. Like Ansible's like all I mean, that might be a bad example. Like, you know, Ansible's been around for so long, but it's just it's it's all about motivations and like what the tool is and like what it's gonna be used for. There's a lot of social media or social networks out there that you can go and chat with people on. Are you gonna go to the the social media platform that's got three users, or are you gonna go to like Instagram or LinkedIn? I obviously you're gonna, you know, and so that network effect that I just talked about almost creates a market dynamic and a market effect where it's just like an avalanche. So you got the community, you got the vendors, like this wheel is spinning, but they monetize them.

unknown 17:04

Yeah.

andy lapteff 17:04

The real question behind my question about Terraform was they were a victim of their own success. It was hard for them to do it because the free tier was so good. Well, right, how many hours did those developers spend building a product that they can't charge for? And why would a developer who can charge $200 an hour on picking a number for their time that they'll use to buy food for their family, why would they do that for free?

William Collins 17:26

Here's the thing, right? In order for something to become important, people have to use it. And this is why, I mean, another effect of this, even with like closed source solutions out there, a lot of times vendors will bite some cost to enable consumers to have a free tier to kind of like play around on so they can understand what it's doing because they want to get traction. They want people to know, okay, with these five clicks and this thing that you run, you can have X, Y, and Z free. Like you can try it out in your own sandbox. I mean, I have burner email accounts from trying out so much free tier. Free tier is great.

andy lapteff 18:00

But that assumes that you can't charge people for a good product. Let me give you like a real example. My nephew works in financial services, they pay $15,000 a year for leads that convert at 1%. It's a cold calling list. So you call a hundred people, you get one, you book one meeting, and out of those meetings, they book X percent, and that's how they make money. They can map it out and know what they're gonna make. So I'm talking to him about this, and I'm like, wow, it's gotta be a better way to curate, you know, higher quality leads. So I spend a few hours and I put a thing together, and in two hours I had it working and we ran it. And I don't know if I'm being gaslit by the LLM, but it thinks that at a minimum, based on the intelligence that we baked in and all the API data we pulled it, like it makes sense to me that this should be a minimum five percent close instead of one. So if I can spend three or four hours right developing this for him and create a 400% increase in their booking rates of these leads, and each of those meetings converts to a dollar amount, that's what I've built for him. Now I've worked it out with him that they're gonna try it. And if it in fact books at five or higher, right? Like if it's worth it, great. And I told him, like, dude, I have limited time, I can't just be your developer like forever. So we got to figure out a way to pay. He's mine, he's family. So like I don't want to charge family, but like there's also time. I'm working basically two full-time jobs between my day job and like the podcast. So like he wants me to work a third job and be a software developer. Like, dude, but the point being is this private repo I have that I'm gonna push to the public place soon so they can consume it. I think I'll get a percentage of the delta between the 1% and whatever this thing generates forever. Like my point being, in a specific case, I could make that repo public and maybe it will be once it's in the cloud. I don't even understand this stuff, right? But for me to open that repo and give that idea away, now it's probably not the first lead gen software somebody built. But if the thing I built gets them a 400 or 1200 or 2000 percent increase in sales, to me, there's value there, and I spent a ton of time working on it. Do I understand the code? No, but it doesn't seem like today you have to with the tooling, at least to a certain extent. So again, I keep coming back to I don't know why the Terraform people spend a bajillion hours of their free time giving it away because they're still are they charging for support yet?

William Collins 20:26

Well, I mean, they didn't have I mean, HashiCorp didn't start out as an empire, it started out as you know, Mitchell coding vagrant and these different things. Like how does HashiCorp make money? Is it do they make money with Terraform? Yeah, yeah, they have a whole suite of products, but basically the whole play here in this sort of market with this type of tooling is to have the open source thing that kind of gives you here's the difference. So it kind of gives you the core of the tool and like what it's gonna do. But there's a difference between writing software and using software that you can use just on a few little things versus like a fleet or you know, an enterprise. Like these are two very different things. Yeah.

andy lapteff 21:05

So anyone who needs leads, like lead gen, yeah, and there's companies out there charging a lot of money for crap leads, apparently. Yep. If I open source that, how do I become hash e corp or how do like right? Like but I don't want to be hash e corp. Like that's my point. That's the thing. You've done two different things here, and it's actually very important.

William Collins 21:23

So you you actually have a real person that's gonna try to use it for real things. This is the the important part. That's when you determine if it's actually useful or not. Well, it might not work, right? Yeah, but I'll tell you the one thing with these LLMs is they're overly optimistic. They're very up. I mean, it's like they just trained on John Capobianco all the time. I love you, John. You know it. Well, that's what I meant by optimistic person. Like minimum five percent. But I'll tell you one thing that I do with my when I have an idea, I have a skill that I've set up that basically it's I call it feature scout, and it scouts an idea. It can be a feature, just any idea in general, but it looks at market dynamics. But one of the things that I have in this skill is treat this idea as if you were a Soviet era Olympic judge. And I mean I'm serious because then you get the real you get the truth. Right. It's telling, it's not pulling any punches. I did that once with Chat GPT, like be my relentless mentor, dude. Uh yeah, it's tough love, man. Yeah, humans aren't ready for that. But but to

andy lapteff 22:16

Your point, like, yes, make me better. I don't need someone to compliment me, right?

William Collins 22:20

Like, I need you want it, you want something to be good enough to where you can sell it and and make some extra cash.

andy lapteff 22:25

You just reminded me, I don't know if it's related, but I saw something that you can take your product or the idea to an LLM and say, assume this already failed. Tell me all the reasons why it failed. And it's a really cool kind of reverse engineering way of seeing those gaps. Instead of releasing it, waiting for it to fail and having it fail, assume it failed, and then it'll pick out those gaps or things you haven't thought about. I haven't tried that yet. I probably should with this software.

William Collins 23:19

Yeah, it's crazy how this stuff, because I I've been working on a tool for a while too. Um, it's like uh basically an MCP tool that takes it looks at your system and sees what L you know uh clients you have, like Cloud Code, Claude Desktop, and all the different ones. Because I test all these things all the time and it actually instantiates the config and creates a single it manages the config for you, so you don't have to mess around with anything. Saves me a lot of time. And then you can stack it's almost like container lab for MCP servers. And I'd written this thing locally on my machine as a CLI tool. It was garbage, like only for me because it was a problem I was trying to solve.

andy lapteff 23:54

And then what does it do for you?

William Collins 23:55

I so imagine like container lab builds and instantiates all these nodes in a very repeatable and ephemeral way. It's like disposable infrastructure in a sense. Well, what this tool that I built does is it does the same thing for like MCP servers and all the surrounding things with MCP servers. Because I do a lot of demos, a lot of prototyping, and I need to basically rebuild exact environments that I'd built before for different reasons at different times. Yeah, it's you know, and that's kind of a mess when you do that yourself. Like there's a lot of just manual updating this file, that file, container lab for MCP servers, pretty much, yeah, which doesn't exist apparently because you build it. Yeah, so actually somebody convinced me to make it open source so they could use it, which then I've actually been moving the project to AI and like using AI to do more of the work now to you know it's not like it's a terraform or anything, so why not?

andy lapteff 24:45

It's a really good idea, right? But I wouldn't fork it like so. I'm I'm pressure testing my own, you know, thought of like someone stealing the idea and going. It sounds like a really good idea, it sounds like it doesn't exist, and I'm also not going to fork it and try to make it mine or try to out like develop you. So do people fork software and try to make money at it and say it's their idea?

William Collins 25:08

But I think the the thing is the difference there is when you fork something that's already built some measure of a moat, it's meaning people are using it's already been proven to some degree. That's one thing, and that's what most people try to do because that's what you know worked. Like this pattern actually worked. How can we rebuild this pattern in a different way and package it in a different way to have a different set of value criteria that we can actually sell and market? Whereas if you're looking at projects that have like a hundred forks or something or a hundred stars, I not many people are gonna fork that and try to build something out of it, probably. I mean, most likely. Yeah. It's just one of those things. The market and the developers, like developers are really good at finding things that allow them to cut time and make their workflows more optimized. They find the good stuff. It just happens.

andy lapteff 25:57

And I'm assuming, again, I'm I'm I'm trying to surface my own biases and like assumptions. I am assuming that no one else on the planet has had the thought that we need better lead gen software, right? Like, I mean, that that's a ridiculous thing.

William Collins 26:09

A lot of people have probably given up on it, to be honest. It's a it's a tough, it's a tough area. What I was gonna say I had a point that I was gonna make about that. Oh, you're you're doing the right thing. That's what I was saying. By by getting actual data to test and prove things out. But what I wanted to say is that there's a difference between making something publicly consumable and making the code open source and available. So say it again and say more about that. Yeah, so the repo and the source code don't have to be public. But you could build a product that you know, you you have a build pipeline that you build and it builds this cloud app, say Andy's lead gen software app, and that's available in cloud to consume. But people aren't worrying about they don't see the source, they don't have to do any local building. Like you're almost you you have the hosting mechanism too, though. So you're taking on more responsibility. Right. So there is that part of it as well. But I would have to figure out how to create user accounts and get all the things and like whatever.

andy lapteff 27:11

Yeah, yeah. But that's a business model that you could do, right?

William Collins 27:14

Exactly. But that keeps that kind of like scratches a few of your itches, but also gives you a ton more work where your actual IP, your intellectual property, still remains private in a repo. Nobody can see it and replicate it.

andy lapteff 27:26

I like my brain is exploding of what you just said, and I'm trying to internalize it.

William Collins 27:29

I'm gonna explode it a lot more here in a second.

andy lapteff 27:31

Well, in my mind, making a private repo public is the same as publishing it publicly, but it's but it's not. I think in my mind, people are gonna steal my idea and fork it and do like the fur I lose the first mover advantage, right? So I like my plan is once my nephew's company uses it for six or twelve months, if it's half as good as the LLM wants to gaslight me to say it is, like if it's worth it, oh okay, this does work and we iterate and we make it better, then maybe I could offer it, you know, publicly. Like, hey, here's this thing I'm doing. We got some data, like check it out if you want.

William Collins 28:09

Yeah.

andy lapteff 28:10

Um, but I would still keep the repo public, right?

William Collins 28:13

Or I'm I'm sorry, private, because Yeah, I mean, you don't have and that's the thing, I would wait for the data to come back to maybe do anything.

andy lapteff 28:19

You know, that's where it's at because I'm always looking like a hundred I don't want to make it public if it's a big dumb mistake. So let's pivot to what happened to me this morning. And well, well, no, no, no, no, like we'll we'll keep names out of it, but the first GitHub repo that I made public, like somebody took a shot at me. Because I guess when you use Claude code to code and you push it to GitHub, Claude makes itself a contributor. I didn't even know this.

William Collins 28:48

But oh, like a public contribution, so it'll show your name and then Claude's name or something.

andy lapteff 28:53

Well, so this very like long story short, I I I created a new website for the show. I I put more value in it, so it's not just hey, listen to our episodes. Like, oh, here's some like useful stuff that people could, you know, could consume. And then I'm like, oh, well, you know, I'm I'm doing this coding and like clause, so yeah, I should put it in GitHub, and then we'll push it from there over to Cloudflare. Like, so I'm I'm I don't know if that's even a pipeline. I don't know what I'm doing, but I'm trying to like get the tools working to and then I started looking at GitHub and I'm like, oh wow, right. Like, I people could, if there's a dead link, or you know, you could report an issue in GitHub and I could fix it. You could do a pull request and suggest a fix that I can merge. So, dude, six months ago, I didn't know what any of this meant. And now I'm all excited because I'm like, oh I love this by the way.

William Collins 29:36

So this is a world this is a world I've seen a lot of people struggle with in network engineering specifically. Yeah, and Claude is well, not Claude, the LL like LLMs and the way that they've been packaged and provided have been a great learning mechanism, not just for network engineers, but for everybody. But it bothers certain people, it does bother certain people a lot.

andy lapteff 29:54

Six months ago, it bothered me. So this isn't me being judgy saying someone hurt my feelings and I'm gonna get them on my show. I'm not mentioning names and that's not what I'm doing. Six months ago, I was talking to Erica Dietrich and yelling into the abyss myself, saying vibe coding is bad. You need to understand the code that you're generating. Now I'm a hypocrite, basically, because I am able for the first time in my life to generate software that is useful, that is differentiated, in my opinion. And some of it, like our website, I can make it public, people can help improve it. Like we have been talking about the you know, A1 community forever and how we're here to help. And like, well, now the resource that is the website, like they can contribute too. So it's not just us trying to figure out what's a good idea. Well, I don't know, but I'm trying to make use of the tools. But the the point being of the first repo I make public, when you use clawed code, the contributors to the repo, it has my name and then clawed under it. So somebody saw that. I announced it as um we released the new website, check it out. It's got a bunch of cool resources. Okay. And somebody said, like, we question mark haha, and then screenshotted my GitHub username and then clawed under it. Now, I took that to mean like Yeah, that sounds negative.

William Collins 31:13

It sounds like you're getting it does sound like shots fired a little bit.

andy lapteff 31:16

Yeah, it was. And the we I was referring to is the podcast. Like, we have a new site, we have a repo, you can contribute to it. Like, we as a community, the thousands of people that have been here for six years, like we can make this better together. And I'm pulling along the TradnetOps people who, like me, don't understand code, don't like GitHub, don't understand what an issue is in GitHub, don't know what a pull request is. Like, now I'm like, oh, I get it. I and I want to pull them along. And when I spend a week of sleepless nights doing this, figuring it out, publishing it, and releasing it to the community, and a well-known person in the community points out the fact that Claude is a contributor to it and gives me crap for it. Now I called them out and they backed off and whatever, but I guess but I'm gonna have to go search the interwebs. But the reason I'm mentioning this is so we're talking about private versus public GitHub repos. Yeah. And all the reasons you would do either, both, none. And the first repo I have that I make public, someone takes a shot at me because it shows Claude as a contributor. So what what I'm seeing in in our industry is like there's this well, there's this new tool that gives people the ability to do things, and I guess it bothers people. And like, I mean, I get it. Like, you know, like what do they call it? Like AI, like not slot, but like like I don't know if the code's any good. This is why I am gonna spend, oh my god, I don't know, 18 hours the next two days in this coding boot camp because I am trying and I have tried and I keep hitting my head against like I get stuck somewhere, and I'm like, uh-oh, the seven terms just got lost in my head. And what was the string to the thing to the dictionary? And oh god, now they're doing something with it. Uh-oh, now we got functions like rut row. So I get lost in all the levels of abstraction at some point, and maybe it's just the way my brain works, but I'm trying. And for somebody to take a shot at me for succeeding at something that I don't fully understand, but it's working.

William Collins 33:10

I saw the site. A site looks great, it looks much better, it's cleaner and just a little easier to follow. You know, it's simplified ideas. There's people building like very successful businesses, like software businesses using Claude Code, right? Well, if you look in every major software product today, you'll see a ClaudeMD file and contributions from different AI uh type things.

andy lapteff 33:28

I guess the industry is changing and it's moving. And I just I don't know. I guess what I'm trying to say is it felt mean and hurtful to me that someone who is trying so hard to learn development and is doing it publicly and is being super vulnerable and showing all my failures and warts and all that stuff because this is hard, and I want to normalize it's hard, and I want to bring those 70% of Tradnet ops folks who aren't automating anything that NAF points out at Autocon. I want to bring them along and show them the the the art of the possible, right?

William Collins 34:04

Yeah.

andy lapteff 34:04

And to have well-known people in the community working at some of the biggest places in the world take a shot at that. It just, I don't understand, right? Like it just why wouldn't this be good for us? And maybe it's not, maybe these are my biases.

William Collins 34:19

But anyway, I think one, there's two things, right? I gotta stop you real quick. So first somebody has to stop. You can't take any of this stuff seriously ever, because even no matter who it is, listen, you just have to do you. I know. And but the second thing is there's a level of like human to this that's that's interesting. I read a long essay about this. It's funny enough, it wasn't about AI. It was it's a little older. But the point of this essay was say I work for a big box vendor or a pizza shop. We just throw out boxes, we're just pushing out boxes, we're a pizza shop, we sell hardware, and I work on some part of that software stack. Like, say it's a closed source proprietary piece of that software. And I don't have skills outside of that thing, maybe some industry skills a little bit, but I most of my time's been sunk into getting that product adopted. Well, I believe it's a natural human instinct when you see all these things that are circumventing everything I've worked on and everything I've built and everything I've done for like 15, 20, 30, 40 years for some, and just making it easier, and people are starting to DIY things, or it's just opening up different doors that is going to eventually hurt the moat of my career. And I think it's human psychology to say that there's a, you know, there's many among us that are going to get defensive, but there's a I think there's a difference between being defensive by yourself and trying to figure it out, and being defensive and also trying to attack people publicly. I'm just not, I'm not for all the public stuff. Like, I think we should all be adults online. I've almost a few times in the past tried to sink my teeth into a few things that were just so over the top and crazy, but I I tend to hold back because it doesn't do any good. I get it, like you know, whatever.

andy lapteff 36:50

Like it doesn't matter and who cares, and like I'm doing me and I'm having a great time and I'm succeeding. Yeah, as long as you're moving, that's all that matters. But this one cut for some reason, and I think it's because of who it was, yeah, and who I thought they were. Um, and you bring up a good point. I don't know if it's a bias, like sunken cost, it's it keeps coming into my head, but yes, like we all have spent an enormous amount of time, money, and effort learning the minutia of the stuff because that's what was valuable. It seems that now intelligence is being abstracted away into these models. That thing that made us valuable and differentiated, and the the ability to have a career based on that, I don't know if that's threatening.

William Collins 37:34

Yeah.

andy lapteff 37:35

Right? I mean, it's like I'm working in product marketing, and they're saying like marketing's one of the first jobs is gonna disappear. So, like I hear that all the time, yeah. So I'm kind of like, I get it, but I'm but I'm not like yelling at people.

William Collins 37:46

That was pre-AI too, though. So if your company is going through hard times, are you gonna lay off your engineering team or your marketing team?

andy lapteff 37:52

But I can tell when I read something if an LLM wrote it. M-dash is an actually everywhere for anybody using it. Remove the M-dashes actually in emojis because I won't read something that is so obviously LLM written.

William Collins 38:04

There's a look and feel to it. There is for sure.

andy lapteff 38:07

And I'm a writer, so it just stands out to me like, you know, but I wouldn't publicly trash someone working in marketing who I see used AI for a thing. And that's why today felt different because someone in my industry that at least knows me enough to take a shot at me gives me a hard time because I use clawed code to make our website orders of magnitude better, prettier, more useful, more helpful to newcomers, people who are in. Like, I thought I was helping the network engineering community, and someone who's in that community and like a leadership like known position was like, ugh.

William Collins 38:43

You know, that's not much different actually than hiring a consultant to come and do all this for you because they're just gonna do a straight shot revamp of your website, and then there it sits till you've hire that consultant again to come and do stuff for you.

andy lapteff 38:54

So I do I don't want to spend any more time on it, but it just like there's a massive amount of like uncertainty right now.

William Collins 38:59

And I will say this like I feel angst all the time because right now we can't build enough data centers in the US and elsewhere to actually support the capacity for training and for these models right now. It's a big problem. And more people are adopting it like hand over foot right now, and it's just going to get worse. Token cost, I mean, I I predict they're gonna get really expensive. You're gonna see the free tiers shrink, you're gonna be paying more. And and so there is an element to this that I went, I don't know where we're gonna be in five years.

andy lapteff 39:29

I'm at max already. Yeah, there you go. In two months, I went from free to the $20 a month to the hundred a month because I kept hitting limits, and I'm like, I'm trying to build stuff here, and and I know that the tokens are gonna get 100's not gonna be enough. And like give you a little taste.

William Collins 39:42

Well, it's great. Even that little taste, if you get everybody on the planet a little taste and they're always under that limit, it's a lot of money, it's a lot of capital. And that's why they keep all these companies are IPOing now because they just don't have the the assets to continue building and doing things. So I don't know where we're gonna be in five years, but I do know I I have a true belief that even if you're not an expert at everything, the folks that know how to use these tools to get better results are gonna be probably ahead of the pack in five years.

andy lapteff 40:12

Well, it's like the snake eating the tail. Like, oh no, AI looks bad for a lot of different reasons, but for me to stay relevant, I gotta get good with AI. So, oh no, AI is getting better at like, right? And it's always this way in tech. You have to, you, you, you have to be on the edge of the new thing so that you're relevant. You all went to cloud and I didn't, and I paid a price. You all went to coding and python and I didn't in automation and I paid a price. And now I'm like, oh, this stuff is awesome and I get it, and I can do a lot of those things with this. So I'm all in, but I also I think they're going for AGI, which is going to render us all useless.

William Collins 40:48

And I don't think so. I I don't buy that.

andy lapteff 40:50

You don't buy they're going for AGI. Well, I think $192 billion this year.

William Collins 40:54

I believe AGI is in some measures already here. Like I believe it's already here in the it depends on how you find define AGI and super intelligence. But there's just I mean, hey, our our financial infrastructure, all the ATMs and all this stuff, COBOL's still well, right. So you you kind of like start peeling apart that argument pretty quick. Somebody for some things it will wipe it out. I'm not saying it won't, but I think the job loss narrative is I don't know. My my uncertainty is when robots come along, and if those get good enough paired with this stuff, I don't even know if I want to live in that world.

andy lapteff 41:27

Well, you you stole the you stole the next words from my mouth. So Mr. You know who just pushed all of his money to robotics. That's where they're going. Like I thought the trades were gonna be the safe place, and I would tell my kid to be a plumber, but now they're gonna have robots.

William Collins 41:41

I still think trades are gonna be very safe for a long time. Well, it's it's a supply and demand imbalance. So as long as the thing is, tech is so saturated now because everybody went to college at a certain time, everybody wanted to be a developer or an engineer. Now you have a huge or you have a market and everybody's a developer or an engineer. How old are your kids? Uh, 10, 7, and uh 18 months. I forgot you had three. Mine are eight and eleven, and like I don't know what I'm gonna tell them to study, right? It's a tough one, yeah. It's funny because my son is pretty perceptive on a lot of things. He's picked up enough in the conversations I have with my wife about tech to where he doesn't want to work in tech, even though he likes it, he thinks it's cool. Um, he's like, I think I want to do something else. And I'm like, uh, does it really look that bad at home from the outside looking in? You know, part of the thing is we're we're sitting all the time. Yeah. When you sit all the time in a room, and a lot of times you're skipping breaks and lunches to keep working. You know, you kind of get a little crazy, you know? It just happens. You know, take walks, folks, get outside, dude. See the sun, get some vitamin D. You sound like my therapist. Andy, tell me about your day. No, you're right.

andy lapteff 42:49

That take get good sleep. That's A for me. Yeah, that's the hard one for me. It's hard. Get good sleep, take breaks, which I don't do.

William Collins 42:56

And I forget the third one. That's why I wrote it down. But take walks, get outside, get the vitamin D. Sun is so good for you. That's something I started doing a few years ago. No matter what, I will just get up and take a walk outside, no matter where I am. I do it all the time. I even have a little tan, and it's not even in the dead of summer because I take walks.

andy lapteff 43:12

My secret to a happy life, which I'm failing at most days, is to get to bed at a reasonable enough time that I can wake up at 5 30, 6 o'clock in the morning, get out for a run, the sun comes up, the sun hits me, I get my blood going, and I am like 150% Andy. If I am up late because that's what I do, and then I don't sleep well, and then I don't run in the morning, and then what like I'm chasing coffee all day, I am like 42% Andy. Exactly. Yeah, you know, and it's all about getting to bed on time, and I have a hard time with that. So I found this helpful. I don't really think that we solved, and that wasn't the point to solve it, but people like me can now code. There are there's a lot of tools that can help you. I think you're supposed to be putting them in repos. You have to make a choice on whether it's public or private, you should know some of the ramifications of that or the consequences. I really like what you said that I could keep the code private, but offer the service publicly. And I mean, I don't even have the time. I don't know how the hell I would find time to do this. But if my nephew tests it out and it's great, then maybe I offer it. Or maybe I don't. Like honestly, because I don't have time, if they can just cut me a percent or two of, you know, if this thing works, yeah, and they can just cut me off for the rest of my life, it's just recurring revenue for me. And I'm like, great. You know what I mean? I spent, I don't know, 20 hours developing and deploying this thing. It's useful for them, and they don't have to pay 15 grand anymore for those, you know, crap leads. So that's kind of where I'm going with this. I doubt I'm gonna make it a public service and charge for support. And like, yeah, I don't want to.

William Collins 44:44

Well, I would say one thing too, uh kind of not a not a ding at open source, but just one of the hard things you have to think through. When it's public and you have people contributing to it from outside of your circle and it does things, you have to worry about licensing. Like the legality is somebody pushing changes and you're gonna merge it into your main branch that are something legal that could get you in trouble. So you're, you know, as the maintainer of said project, you have the licensing that you have to go through and choose. You have to understand what that licensing means. You have to have a license agreement for the people that are contributing so they know what is accepted and what is not. You have to have a code of conduct, all these things.

andy lapteff 45:24

Actually, I think open source also susceptible to that? The people working for free developing software have to track all that too? Oh, yeah. Why are they doing open source? That is so much liability and time and awfulness. I guess they.

William Collins 45:39

The bigger the tool, like the more important it is, the more visibility, just like all things. You know, so if it's just like me and I have this cool little project and three people use it and something happens, oh, no harm, no foul. But if you have something that's getting mainstream adoption, and that's when it's usually if it's getting mainstream adoption and it's open source, you don't even have to ask. You're gonna start getting a lot of contributors. People are gonna be helpful because they they're using it because they like it, they see the utility of it and they want it to grow, they want it to expand, and usually it has something that they need it to have, and it's easier for them to just go through and add it because the product already has maintainers and all this other stuff. Like being an open source maintainer, it may sound cool, but I've talked to so many people over the years, and it's it's a hard thing, and it's usually driven by people with conviction that want to do the right thing and want to put in a lot of time and effort and aren't expecting any sort of pot of gold at the end of the tunnel a lot of times. I got a way to end, I think. Is Linux Torvold's rich? Probably, but you'd never know it.

andy lapteff 46:40

Right. So he I know him from like his desk, his pictures of his desk.

William Collins 46:44

He's like in flip-flops and khaki shorts, and his desk is like what mine looked like 15 years ago. But when I think open source, that's the first guy that comes to mind for me.

andy lapteff 46:53

And I guess what I'm wondering is a guy who everyone knows, the whole world's running on his operating system, everybody's using GitHub that he created.

William Collins 47:01

He created Git, not GitHub, so he actually created the protocol, yeah. Which is even better, right? Yep, even better. Accolades, baby.

andy lapteff 47:07

So I'm curious if he's broke because it was all open source and the world's using it, or it got so popular that he was able to monetize it. Like, I is there a support version of Linux that like you can pay for and yeah, but that's right, Red Hat's shtick sort of like um using it and enterprise.

William Collins 47:27

Linux never got adopted as like a desktop operating system, which now it kind of is in a lot of circles, especially Amia is kind of driving a lot of cool innovation around that. But it's running on like every server in the world, but like yeah, the actual like things that run things that are really important, Linux, like even like Linux instances are essentially like load balancers and cloud. A lot of things are built on just native Linux that do a lot of very cool and unique and creative things.

andy lapteff 47:51

And I don't think he'd ever come on the show. I haven't reached out to him yet, but I'd love to know like, did he do all of this just for the satisfaction of knowing he made the world a better place?

William Collins 47:59

Or I need to solve a problem at the time when he built Git specifically, the way that you'd have multiple folks contributing to a singular code repo was very messy. Oh, sure. And I mean it was like the whole Linux thing too. You had these big paid versions, like Unix-based things that you had to give up an arm and a leg and a blood sample for just to use. And he's like, this should be free for the world. And so he he's like one of those folks that is really probably did a lot of that just with his convictions, and that's you know, satisfying to him, right? Whether he makes money or not. Did he license Linux? You mentioned licensing earlier, I don't really understand. No, licensing's different for the so the software that you build on top of everything, that's where the licensing comes in. But Linux is software, it's an operating system. So yeah, they would have a license. All the public anything public and open source that they use is gonna be licensed. But it's gonna be a license, more so there's different licenses, right? There's like an open source license you don't pay for. Yeah, well, there's a lot of open source licenses, like everything from like MIT license, which is very open, kind of like the most open sort of license you can have up to like Apache 2. But license doesn't mean you're making money.

andy lapteff 49:00

Like Linux is or um uh the Linux guy. Linus. Linus Linux. Or Linux, yeah. He's not making money because he licensed.

William Collins 49:08

No, he doesn't make entitlement dollars from from Linux at all. If he would have, he'd he'd be in a palace. He probably still could be in a palace. Who knows? I don't know anything about his finances. Thank you for coming on here.

andy lapteff 49:20

Thank you for uh always a pleasure. Being on the show. I wish we saw each other more than once a year. I'm gonna come out and milk those cows someday and PA.

William Collins 49:28

I'll be there actually next month, I think. We're gonna make we're gonna be there for a few weeks. I love it out there.

andy lapteff 49:33

William comes to my home state every year and doesn't tell me he's there. Two times a year. And then I find oh, and then I find out later he's there, and I see pictures of four-wheelers and all this fun and the best milk he's ever had in his life.

William Collins 49:45

I'm like, ah, one of these years we'll Yeah. My in-laws have a dairy farm. It's amazing in some ways. The more you're outside, like the more you just want to be outside. I know. And then you don't want to be inside anymore. Dude, I love it. I that's my trip to basically I hardly am on any sort of screen or anything to do with tech. I'm hardly inside. I'm outside like all the time, and I love it every second of it.

andy lapteff 50:07

Do you know what I've been doing for three weeks, two weeks, probably? Installing a floor in my attic. Oh now it's summer and it's hot, and you would think it's awful. Dude, I love it. Really? I used to work with my hands. I've worked in construction, I've worked in all the trades. I love being outside, and I love the smell of fresh cut wood, and there's something so satisfying, and which is probably why I like networking because it's tactile, like right, there's stuff to touch and do. And I can't believe that I love being in an attic in the dead of summer, like I'm rerouting wires, I'm moving stuff, all my drywall's falling down. So I did all this gluing and too, but like I've been out there, I'm spending way too much time in my garage, but it is so satisfying working with my hands again. Now, my back hurts, my shoulder hurts, my hand hurts, my knees hurt like it kills you.

William Collins 50:53

I have a wood shop in my basement that I build. I'll show you pictures there, right? I love woodworking. It's like hands-on, you're right. Hands-on is I don't even like listen to anything. Like when I'm doing the hands-on stuff, I just have my thoughts building a coffee table, a bookshelf, or whatever. What are your thoughts? It's just coffee table. Yeah. Coffee table. So smooth, so smooth. Don't think anything complicated. I hate Ethernet. You know. Uh, where can people find you? Uh so I uh host the Cloud Gambit Podcast, part of the Packet Pushers family podcasts, and wcollins.io. That's my blog where I post all my hot well, not even hot takes, but more me stuff. They're hot to death. They're fire. The graphics are hot. You your last episode, I was yelling at my phone. Which means it's good. Was it the one with um the data center one? Oh, yeah, yeah. That one but was really good though. No, like I Yeah, there was no planning in that. Avon and I were just chatting and we're like, this is important. Let's just jump in and start talking about it. So I was yelling at Avon.

andy lapteff 51:53

She didn't hear it. It was just I was yelling at my phone. Apparently, you guys can't hear me when I'm yelling at a podcast. It's probably a good thing. No, but it was no, it was a really good conversation. You should definitely check out the Cloud Gamba podcast. I love Avon, so I don't want to say the wrong thing and get her mad at me, but I thought that it was interesting that knowing what we all do for a living and knowing what part of the country she lives, or you know, the country that she lives in, and that she was having a hard time understanding why locals wouldn't want data centers. Now she made really compelling arguments and the car factories and stuff. So like it it made complete sense, but I really like the episode because it's like, you know, it's like the the Claude co-founder talking about how great LLMs are, like, well, of course, or like, oh, you everyone should use more tokens. Well, yeah, duh. That's how you make money. So, like the the data center person saying, Well, I don't understand why nobody wants data centers. We need more data centers that work.

William Collins 52:49

I think the longer, the longer you've been around the industry, the more you're probably okay with that, to be honest. Because I'm I have my own takes on that, which I don't know if I really made them as I was honest, completely honest with how I felt in that episode. Um, I kind of feel in the same vein as Yvonne. Like, I understand why like people that don't understand what we do, they're irritated about it. They're irritated because there's so much misinformation out there. I'll start with that. Like, I see so much, even like targeted stuff. Like, there's gonna be no more water. We electricity, this, electricity that. You're gonna be in the dark. Like, there's a lot of misinformation, even if there's certain data points that are true, it's important to frame those data points correctly. Like, you just don't want to be a hundred percent fear mongerer. Like, you wanna be a little more exploratory with how you do it. But with our media, like I can hardly read a news article anywhere anymore because they are all just clickbait that are trying to frame some narrative on every side of the fact. Even the sides I agree with, I hate a lot of this writing because it's just the way that it's done and the way that it's presented is so disingenuous. Like Avon makes fun of it. Well, like Avon will seriously tell me, like, William, you need to check out the news. Are you in your your time of you know, your time that you're in? Because I just I'll go in these like rabbit holes for two weeks, or I just won't look at a single thing because I can't I gotta get my brain away from it because it's so dishonest. So the news has always been bad.

andy lapteff 54:13

Um a hundred years ago in college, I had to watch the news. I I had to watch the news every night for a semester. I was a communication major and we were studying media, and I was, I think, like close to clinically depressed by the end of the and this was this was a hundred years ago, because the same reason we all stop and look at a car crash is the same reason that the news like they feed into this thing of like chaos and conflict grabs our attention. And and I I think it's just I I had a thought as you were talking, like I think that AI is very threatening to a lot of different uh people and places, meaning what we were talking about earlier that those people who spend a bajillion hours getting all the certs that are threatened by these models that can do things better than them now, it's almost the same threat of these local folks who are like, Well, I don't want a data center in my neighborhood for reasons. And whether their reasons are founded or not, we're all in this interesting disruptive time where we're pouring more resources and money into this that has ever been poured into anything in the history of humankind. And there seems to be a relatively good chance unexpected negative outcomes could come of it. So I don't necessarily know if the people that are worried about their water table or pollution, like I don't know how founded that is. I'm not undermining what they're saying.

William Collins 55:40

There's an element of truth in everything. So anything that there is probably some element of truth in everything you read, but it's just and I think you guys touch on this in your episode.

andy lapteff 55:48

So that's that's where we can finish it up. I don't want to start a different episode. We got new people coming in 10 minutes, but what is it? We can't put the genie back in the bottle. Yeah, correct. And all the founders of machine learning, AI, LLM stuff have all quit their jobs at places and gone on record saying it's really bad what's happening, there's no safety or guardrails, we're gonna end humankind. And these are like triple doctor, like these aren't idiots. Like the people who created this stuff because we're all chasing financial incentives, right? Like we're we're running toward the dollar. So I think that they're all related. I think that the people threatened by it because of what they've learned is being maybe replaced by something more intelligent than them, the people who are worried about their impacts on their community, and all of us who are like, what does this mean? What am I gonna tell my kids to study? Like, there's so nobody likes disruption, right? It's like a very I find it to be a particularly stressful time in the world, right? And I know it's the world's always on fire. I try to remind myself of that. Like, I'm not a religious guy, but you can read a book 2,000 years ago and the world was on fire then. Yeah. So like I think the world's always on fire, but this just feels like if different, if we don't know what to do for a living or careers or sustenance, or to like take care of ourselves and our family, which is what some of this feels like. Like, I love what I'm doing with the tool, I love what I'm seeing in networking, troubleshooting and building, like it's a great tool, but I'm not sure how long my job will exist if this stuff gets better and better as it is. It's like Moore's Law, right? It's getting, yeah, it is improving dramatically. So for people to say, Oh, this isn't really going to cause disruption, I don't know how you can look at the incremental progress every what is it now, four months?

William Collins 57:34

Like, I don't know when these moments are pretty constant because they don't release at the same time. So each new frontier provider seems to have a different cadence because they want the marketing ground.

andy lapteff 57:43

So when so when, oh, we got the next come on in, neighbor. It's it's Mr. Rogers neighborhood. There he is. What's up? Come on in, guys. This will be great. We have the two parts here. Guys, say hi, come those parts. These are our next guests. I'll wrap up the show and then we'll start your show. Heck yeah. Um, watch your watch your step with all the so to wrap up the conversation. I really enjoyed this. I think we touched on a lot of different things which we usually do. We could probably be here for three weeks and keep talking about this stuff, but very good topical things that I think apply to all of us. Um, and um, you're the first episode today. We talked about AI. So hopefully the last you're probably all gonna touch on it now. Yeah, yeah, wink wink. All right, so people know where to find your cloud gamma podcast. And what was this site? Uh, wcollins.io. Just a blog.

William Collins 58:30

Just a blog. It's a fantastic blog. I'll give you all my links.

andy lapteff 58:33

Yeah, the links will be in the show notes. Uh for all things art of net end, you can check out our Linktree. It's Linktree forward slash Art of NetEnd. Check out the Discord server and be sure to check out the new hot website with a bunch of cool resources. There's a resource section you can check out. And uh, we are releasing a newsletter probably in a few weeks. So sign up for the newsletter. That'll be out soon. Thank you so much for watching and listening. We'll catch you next time on the Art of Network Engineering podcast. Hey folks, if you like what you heard today, please subscribe to our podcast and your favorite podcatcher. You can find us on socials at Art of NetEng, and you can visit Linktree forward slash Art of NetEng for links to all of our content, including the A1 Merch store and our virtual community on Discord called It's All About the Journey. You can see our pretty faces on our YouTube channel named the Art of Network Engineering. That's YouTube.com forward slash Art of NetEng. Thanks for listening.