All right, here we are. I’m outside, I’m getting my vitamin D in with a post-lunch walk and I decided, hey, I’m gonna record a short video with some thoughts that I have. and I recorded a new episode of Raising an Agent this morning and it’ll be out soon. But I kept thinking about one topic, so I’ve never done this before, but I decided, hey, why not record a short video, right? Short form video content. And... The thing that is on my mind is this question of is GitHub dead? A lot of people talk about this and it’s come up a bunch of times online and I think when people talk about whether GitHub is dead, what they often mean is something like Microsoft doesn’t invest enough in it or the software quality on GitHub is getting worse, nobody’s investing in it, it’s falling apart. I think GitHub is dead, but I don’t mean this. What I mean is that I think a lot of development tools that we’ve used for the last 10, 15, 20 years have been built on an assumption that no longer holds. And that assumption is code is expensive and slow to produce. Code is written by humans. Code is valuable. And you can see this in... you know, let’s say on GitHub, you can see this in a lot of software, but in GitHub, for example, you can see that, well, a pull request is something valuable, right? It’s something that you can leave emoji reactions on. It’s very prominent. It’s a first-class feature. You can react with the heart. You can bookmark it or favorite or assign people. Like code and a contribution to a code base is seen as a valuable thing. And it still is in open source, but fundamentally with agents, a lot of the calculations change, right? And a lot of the value of code changes. And when I think about development tools now, I think they’re all built for humans. They’re all built on the assumption that a human has written that code. That means a human will write this code. That means it’s gonna take a long time to write. So that means I’m gonna create a ticket before I’m gonna assign something. I’m gonna put this in planning in two weeks, all of that. But if you have a legion of agents being able to knock out code as fast as you can basically paste the screenshot in a Slack chat or in a prompt input box, that whole calculation changes. And I think that we’ll see in the next 10 years that most of the development tools that we’ve been using will completely flip. And everything we’ve been using before, I think, is either dead or slowly dying, or it has to completely change the way it works. What we’re going through with code is similar to what infrastructure went through 10 years ago, where, I don’t know how many of you know this, but the idea of a server shouldn’t be treated as a pet anymore. A server is cattle now. It’s not, I don’t know if there’s a cow here somewhere, but a server is cattle. And prior to this shift in infrastructure where we had infrastructure as a service or cloud and. You had Terraform and all of these tools and Docker. Prior to this, you would give unique host names to servers. You would maybe list all of your hosts or your servers in a config file. And now, who gives a shit about which server something’s running on? The server is abstracted so far away that you don’t care about the host name anymore. It’s not a path anymore. part of your big herd of cattle. And I think what we’re going through with development tools will be similar to this. Where the UI, the UX in these tools will change to reflect the fact that code and code contributions have turned from pets to cattle. And that’s also what we’re really focusing on with AMP right now and what I think is the next big step. That’s what I mean when I say the system is dead, long live the factory. Yeah, let’s not get into cattle and factories, but yeah, there’s a lot to build. All right, that’s it. Let’s see how this goes.