Open Office Hours

Welcome everybody. Today we got Ray with us and we got the software veteran. I forgot the term, but you're the software expert. We don't have a topic as always, but we're just going to talk about software and also something, a new opportunity that is coming up for many people to benefit from Ray's and my, my knowledge. When it comes to building applications, you want to talk about that? Because it pretty much was, you know, I started the idea, but you took most of the initiative.

Yeah, well, I mean, we've talked before about how, you know, we say we're going to talk about software here. And, you know, it used to be like software was something you hired someone else to do, right? And now there are more people who are able to go build it themselves. Different change in demographics, but also change in tooling, which are low-code tools, AI tools, what have you. But the other thing is like what you and I do. Where we have, you know, amongst our other, you know, practices. And I guess we'll talk about one of your other products a little bit later in the hour. We have, you know, I run these communities, right, based on what I call micro consulting, the idea that you don't need hours or days or, you know, you don't need to hand off someone entirely and you don't need like, you know, huge amount of consultative advice. What you need is basically when you get a little stuck, be able to get unstuck and be able to keep on moving. And there are some processes, right, or some mental patterns that we can learn that will allow us to move more quickly. That's the premise of the office hours model that we have here at StateChange. Thank you for the QR code, but also at Low-Code/Pro-Code. You and I have been doing this for people who are trying to build their stuff both on the front end and the back end. We even have a couple of people who are members of both communities. The insight, which seems obvious in retrospect, is that I've got a lot more expertise, right? And people come because of my, you know, expertise around code, around, you know, backend data considerations. And also I'm more of a data scientist by training, which sort of makes it more of a natural fit for the, for the AI part, at least the more mathematical part of it. And, you know, you've got lots of, you know, deep experience and credibility on the frontend, right? Issues of design, right? Issues of UX, right? Issues of how to build and how to build fast. I mean, you were at one point, you know, doing a lot of Wiz work and you were doing a lot of, more recently, Nordcraft work and WeWeb work. And you have some products around that. And people can come to you and they're able to make that kind of really good progress. People sort of sing your praises to me. Anyone who has met you and seen you in that environment can't stop talking about it. And so the notion was that there's probably more people would benefit from having both of us. So we created sort of a gold pass for, you know, people who want to join both for less than the price of each. Because I think like joining both of our communities was something close to like $300 a month. And so we made a $199 offer, uh, that is, uh, that actually if you go to Sageinge AI, you'll be able to see at the top. I don't know if you put it onto your site yet.

Um, I was a little bit busy. I'll do it tomorrow. Yeah, yeah, because if I open my site, I'll, I gotta fix 15 things before I add your thing. I'm just, that's just how I work.

Yeah. Yeah, we got the backlog. Uh, and the, uh, but the idea is that like, you know, not that people should go and join both things forever, but that you go and you get help and you're able to then ship just that much more quickly. And I, I, the SayChange Pro office hours are like 5 times a week. How often are the ones over at, uh, uh, NoCodeProCode?

5 times a week every weekday, right?

So like between the two, I mean, there's like, that's, that's 10 office hours per week. That's a lot of opportunity to get a lot of help. And so what— and so it's an experiment. All right. I would love to have someone sign up for it and abuse the heck out of it. I want someone to show up, show up to as many of the office hours as they can. I mean, I think for both of us, we're kind of spread out for different people around the world. So someone who is in London might have, you know, find that there are going to be, you know, 3 office hours per week, say, from state change that are good times for them and a couple that are kind of bad times for them. Just like someone who's in, say, the western United States would find there are 3 good times for them, and then, you know, a couple that are— and, and like, you know, 2 that are not so good. Someone who's in Asia would find there are sort of 4, what, you know, uh, early and late, uh, that are good for them. And then the one that would— I would call middle of the day is of course middle of their night, and they're, they're not so excited about that. But the, um, but the idea is, you know, something that's going to be convenient for you and that you can make a lot more progress. I would be just totally excited to have someone join and then be able to kind of race through and be able to ship in absolute record time as a result of using the best tools in the world, you know, their own commitment to the project and being able to get that kind of help on both the front end and the back end and sort of be able to ship to the other side. And not like in 6 months, but like in, in 1 month or whatever is required. With that in mind that we made this to see if we can help people go radically faster.

You're Mr. React, you are Mr. Vibe Coding, but you don't just know how to vibe code, you know the actual code that is behind it. If someone says, you know what, screw it, F no-code, I want to make an app in React, but I don't want to use React Native for some reason, I want to use Despia. I want to use some hybrid frameworks, I'll be able to help on that part and Ray will be able to help with the React part. Then I'll be able to come in there and consult you and give you some insights on UX and UI, which you can look at the state change landing page. That is not Ray's strength, but this is where it's the yin and yang. Yes. I tried to do it as nice as possible. I apologize.

Nope, nope. I, I, I, I've earned that. The, uh, yeah, you can sort of tell I am not the, uh, Mr. Graphic, uh, Design. Um, I work in marketing, I work around finance, I work around, uh, you know, like I say, the sort of the deeper parts of technology. Um, and but there— but UX is really important, and that is something that like Yacine knows, uh, super, super well. Anyway, so that was the idea we had, and that was the experiment that we're hoping we'll be able to do, and we'll see what we'll be able to do in the month of September.

Fantastic. And then if you, you know, if you're using WeWeb, if you're using Nordcraft, if you're using anything, we'll be able to help you with the no-code side. And then, you know, if it's more about the analytical stuff, how do I do that? How do I work with this custom Vue component? How do I get a custom React component working in Nordcraft? You got Ray's extensive experience and, and my, you know, round border button skills.

I think anybody who's worked with you knows just how, you know, inaccurate that self-deprecating comment was.

I mean, to be fair, I have people in my community, when I show them something, how I'm adding a button, they're like, let me guess, you're going to add gray 200 to it. It's James, you know, it's like gray 200 is my favorite color.

You have gray 200. And I keep on talking about stop and debug and console.log.

I mean, I do console logs too. We do console logs in office hours, but I did inherit that from you, I would have to say, because I'm like, we just do a console log and then we'll do a try-catch. I was like, my God, I'm sounding like Ray.

Well, there's medication for that. This is great. We'll come back to the Despia thing, you know, I think a little bit, uh, in the hour. But like, you know, having, having talked about one thing that's sort of new for us or our idea and our initiative, what are we seeing that's new out in the market, sir? What, what, what, what?

I saw one very cool thing, and I think this no-code tool may be actually something that is wiping out Nordcraft and wiping out WeWeb. It's called VibeFlow. What it is, it's a note -based programming interface like NodeGraph where you do your formulas with Node-based programming, but it incorporates Node-based visual programming with Vibe Coding. Let me see if I can pull up a video. I did not prepare for this, as you can see, because I've been working day and night on— if you just like look up Vibe Flow, you only see yoga studios in your area. I'm getting like 5 yoga studios in San Diego here. But let me just look, their AI app. So it's a visual application builder. That's what it is. Let me see if I can share my screen so the viewers can get an idea of what this is about. Do you see my screen?

I do see your screens.

Perfect. Let's make that bigger. Okay, so this is VibeFlow, and what it does is it's essentially— you can vibe code an application But then you can go in there and you can work with those notes like you do in Nordkraft to configure the dynamic data and the backend of that application. So it combines visual programming, visual workplace, workflow programming with vibe coding. So it takes the best of code and the best of some of the principles that we are seeing here you see you have this node-based programmer where you have your inset, your embedding nodes, your database, and you can select tools. You can build your Vite coding application, but it essentially creates this visual markup schema that lets you visually build your application. The AI can give you a starting point, but you can then complete it visually, which is pretty darn cool.

It is, it is. You know, the, um, this is actually really interesting. If you were to scroll, uh, first I would also note that, uh, VibeFlow is a, um, it's a Y Combinator-backed company, so they've got venture capital funding. They see— it looks like they're in like the current batch, um, so I'll be curious what fundraising they do because usually that's, that's how the end of Y Combinator works, is that— or Y Combinator, uh, a batch works, that there's like demo day and they raise money and, you know, So this is like the new venture capital back thing. And I look at this and think, okay, my friends at Xano are going to be quaking about this a little bit. But if you could just scroll down just a little bit more, because I think there's something else in here that's worth noting. You were showing that the vibe, the visual part for a second, but if you scroll down just a little bit more here, I think what I'd really like to do is highlight— I don't know if you can zoom in with your, with your screen share— but this live code generation thing. And I want to bring this to your attention because not only is this like, you know, sort of interesting feature of, of our friends over at VibeFlow, but this gets to like a whole way these tools work. And I think there are sort of two types of tools out there. One type is what I would call a runtime. Xano is a runtime. Um, uh, Nordcraft is a runtime, right? These are both that you have to go run it basically on their infrastructure, and they might allow you to go run their software on something else, but like, it's— it, it goes down relatively deep, right? The, the, the, you know, configuration or whatever you're creating needs to be run on the software they're providing underneath in order for it to continue to work. Wizd basically is this. Um, you know, Nordcraft works this way. Um, WeWeb is not this, right? Which is where things get a little bit interesting. Uh, WeWeb, for example, is a code generator. It's in the business of creating HTML, JavaScript, and CSS. It's actually creating a Vue.js app, like Webflow. Uh, Webflow also creates, uh, you know, those, those same three things and can be exported something else. Um, and the idea is that these are code generators, and what they create are code essentially text files that you could then go host somewhere else. You go host it on Cloudflare, on Netlify, on Vercel, whatever. Just go get your website, you know, up and going. When you think about most of the by-coding tools out there, like the way that, you know, Cursor works or Cloud Code works or Convex works, Codex works, those are all, those are all code generators, right? Because actually their business is to go create the code and get the text file at the end, right? We call by-coding coding because it creates text. Lovable. And Bolt, kind of in the middle, there's some degree of exportability, but they're kind of designed to be run, you know, inside their own environment. And then V0 is an interesting play because yes, it's just a code generator, but it has a really close integration to make it easy to deploy to Vercel, because the V in V0 stands for Vercel because it's a Vercel product. Now the reason I bring it up is because of that box you had right there, that box you just had up showed what code you're creating with, um, with Vive Flow, which is to say it's just creating JavaScript. It's just creating a very specific form of JavaScript that then can work back and forth with the, um, with the, with, with the visual play. And I bring this up because they didn't have to go make up a new language. They just used an SDK for a language that exists to be able to make it then just work. And I think that VibeFlow uses Convex, right? So it's a way of like building Convex backends. So the infrastructure they work with is working with Convex and other sort of best-in-class tools that are popular with the coding and indeed with the Vibe coding set. And now instead of you having great code yourself, it helps you.

It helps make things happen. What most people don't understand where WeWeb is a tool that builds a real application for you that you can host. You could technically open it up in Cursor and, you know, try hacking around with it even though it has its own abstraction. It adds attributes, you know, for WeWeb. But, you know, you need to do that. Otherwise, the editor doesn't know what the heck is going on, right? But WeWeb is a, you know, a code generation tool that generates universally usable code that is not bound to a specific infrastructure. Where Nordkraft is, they made up their own programming language, same as Wist, and they have a code generator that only generates their programming language. Language, which the downside of that is if they stop maintaining their programming language, you're at the risk of your product slowly dying off. Where with WeWeb, if they just keep making sure that the editor runs on the latest version of Vue and they stop having components that break your whole application, like was sometimes mentioned in the community, but I think they fixed it. That's all they got to do. Of course, community would wish if they would do more and they're doing more with AI, but they have someone else manage all the dependencies, all the complexities. They don't have to worry about expressions and all that random stuff. They just build on top of Vue and they just have this code generator that generates Vue code where they don't have to maintain a programming language, which is arguably a little bit more work. It's probably better building your product on top of a programming language that is used by actual programmers and it is an actual stable ecosystem on top of someone's proprietary programming language that's maintained by a couple of people, right, as other no-code tools.

Yeah, and I think you— that, that last point you made about maintenance is a really key one. Like, if you're going to work with— if you're going to be depending on something, how much are you— how long are you depending on, and how much, you know, are you paying them to take care of things, you know, for you?

But this is also— I think Nordkraft is going open source because people are looking at that and They are like, this is a, they have their own programming language. Essentially that's what a runtime is. And you know, what if they just go bye-bye? You know, and I think this is the reason why they open source it. So to give people the confidence that should something happen to them, they can keep running it. But the truth of that is I actually talked with the CEO of Momen about that, which is an up and coming no-code builder. They're focused on AI. I see it as a web disruptor, but they're not quite there yet. The CEO told me it doesn't make sense having your proprietary runtime open source because who's going to maintain it anyway? Probably the only person understanding it is me. I was like, that is a good point.

Yeah. People think that open source is the answer to that problem. There's another answer, which is the contingent license. So like, uh, I, I remember when I was— had a— when I was first starting my legal technology company, I had these law firms say, well, what if you go bankrupt? What's going to happen? Like, well, okay, I guess you're just going to get to keep on running this for as long as you want to keep on running this, you know, in the event that like we're not able to do things. And the, um, and so what I had to do is I had to make a license saying, or make it part of our agreement, that they would get source code, right? And more importantly, they would get a license to do everything they need to be able to maintain the software, you know, after they had purchased it from us. Because not every— because we were doing some perpetual licenses in addition to like—

oh, that's a good— that is a good— I got to do this with Despia. Should anything happen to Despia, everybody gets a copy of the CodeMagic YAML file and the Norcraft backend, if Norcraft is still around by then, ready to maintain, plus the Xano stuff.

We'll do this. Right. And, but really what that does, it just puts the pin in the grenade for a day, you know, like, if your vendor has gone down, it is time to switch. And the—

I just want to put this out there. But it's a good idea. Sounds good on the landing page, right?

Well, yes. And like people, and I think there's a little bit of confusion as to what open source is and what it's good for. I mean, the, the, of the, the world's come a long way since Eric Raymond, uh, was sort of making this stuff happen, uh, with, with the GNU Foundation, but the, uh, or in the Free Software Foundation. So I'm gonna say, and he actually— I used to work down the street from him, um, and that was, that was kind of, uh, neat. Uh, I remember talking to FSF about like us, you know, about my company distributing open source software and like the FSF did not like it. But of course FSF didn't like, you know, people being charged for software because they didn't think that's what should happen with software. Um, but, but of course it just became sort of the, the standard, right, that you would like deploy a virtual machine on Ubuntu or something and that was just the way you could deploy it and then you wouldn't have more licensing considerations. But the, uh, but the, but I actually do think that the, that, that's a little bit of a red herring and people are concerned about, well, what if I build on this and then the company goes away? And I, I would, I would suggest two things. The first is it's far more likely that your business goes away before the company, before these vendors do, just because they've got money and they're going to be— even if they go into zombie mode, it'll be usable for longer, right? And then, and then you're asking how to switch over. And the second is this stuff has gotten the, the, the, the evolution technology is making it all deflationary, so it will be cheaper tomorrow for you to go switch over from what this is today to something else, and you'll just say, well, what was I ever worried about?

As long as they give you access to the keystore, otherwise you're effed. There's some no-code tools, like some especially no-code mobile app builders that like keep the keystore so you can't switch to any other one.

Oh goodness. Yeah, yeah, they're probably Yeah, actually, that's a really good point. There are probably some things you do need to be concerned about making sure that these, you know, fly-by-night vendors don't own. But, um, but most of the time, like, from a how much is it going to cost me to develop point of view, once you know what it is you want— I mean, my thing about low-code has long been that, like, low-code really makes the rebuild cheaper, not the build, because the build is usually more about the business needs. But on the technical side, this stuff gets much easier, and that same principle applies to AI parts.

I had one client and I quoted them and they said, how is your quote for building this in no-code as expensive as the stuff we got quoted in React? I said, it's the same work. We will have to sit on a meeting and we will have to understand what your product is about and we will have to build it based on your needs. And we have to understand and do the discovery and we have to think about how to build the engine. And that is essentially what you're paying for. You're paying for this proprietary engine. You're not paying for the code, you're paying for the result. And the result is the same. And I do think my result is better, but I also sound slightly arrogant when I am very confident in my skill. So, uh, yeah, that's what I told them, and it made sense for them because you're not paying for how, you're paying for the result, right? I think that's a result that's more ownable that you can understand, that you, you know, you're not looking at like 50 files or 500 files and you have no idea where something is, you know, that is a better end result. I interviewed one from our community. They're a big enterprise, they sell private jets, and they build a mobile app where you can just like casually buy, you know, in the morning, like while you're scrolling on your phone, a private jet, swipe to the right, get a private jet, you know. Like for $20 million, you know, like that. Um, rich people's problems, uh, that's what he said. And, uh, and they are able to fix your application in a matter of minutes because he says all those rich people, when they buy a private jet right now, they want to buy a private jet right now. So when the button doesn't work, we're losing millions and millions and millions of dollars. So we need to be able to have the intern go in there and be able to just like somehow fix this button without messing up everything else. And that's why they're using no-code. Goodness gracious, it's a problem. I mean, it costs them millions of dollars for a button not to work if you're selling private jets on a mobile app, right?

Well, yeah, it totally changes the, the dynamic of like which parts are important and who you can afford to work for and blah blah blah. Um, but, but, um, but yeah, the, the, the— and that's what I really like actually about, you know, building in this space and helping people is that it's much more— it is, to your point, that the core of it is really still that business consulting style. It's like, what's the job to be done? What's the business? Where's the money to be made? How are we— what's the promise? How do we like link this back into marketing? You get to focus your attention on that without having a lot of, uh, separation or, um, you know, specialization of labor, right? Because now you— it's like the product person interacting with tools as opposed to product person interacting with their engineering liaison who's interacting with an engineering team.

3 days for the engineering team to do anything. They boot up like an old Windows computer, right?

And, and I think an engineering person would, would say correctly that the reason it takes them 3 days to do anything has to do with like writing tests and whatever, because they're going to build it out of stone. They're going to make sure this thing is able to last, or as they put it, scale, or what have you. My general take is that it's not nearly as important early on. This is what we've talked before about like product excellence versus engineering excellence. And the, uh, but later of course it starts to matter more for some parts, and that's going to be what— and that's what a good product owner is differentiating. It's like which pieces are important when, uh, and then that's how you gut the cost from it. Not through the initial, like you say, the spin-up part, because that initial spin-up part is so much about business understanding and, and collection of whatever.

The difference of Once you have this built for, let's say, $50,000, $100,000— there are WeWeb agencies charging people $100,000 to build a WeWeb app, surprisingly. It's not like, are you paying $100,000 to set up a WeWeb app compared to $100,000 to set up a React app? Even though most React apps are now vibe-coded, you know? So I think actually there's more effort and thoughts being put into the WeWeb app or Norcraft. But it's like, okay, paying $100,000 for both, it's fine. But The difference is, are you paying $2,000 a month for maintenance moving forward, or are you paying $20,000 for 2 people doing maintenance full-time?

Yeah, that's a good point.

So it's the recurring costs that you, you know, like building software is cheap, maintaining software is expensive.

That's a, that's also a good point, which brings us back to like, which brings us the difference between, you know, companies that are build tools versus companies that are infrastructure or deployment tools. So like the, the build tools, this would be like kind of where— and, and, and where I think some of these companies get in trouble is when they're trying to be both those, right? The theory is, hey, you come in, you use our nice build tool, and then you're going to stay with us forever because you're using our infrastructure, right? And that becomes the— that, that, uh, that capability.

I do have one rule with Despia. We— because we are a build tool, you are unable to configure a Despia application unless you use our M2 build resources because that's kind of like the whole magic sauce. Of course, we can send you a template of the framework and you fill in the gaps manually and put in your modules and add your targets. If you have a week to spend on that, you can do that and we can send you tutorials on how to do that. But why would you do that, right? The whole idea is you can just ship it to the App Store. So there, there is a dependency on the build tool, but what we don't want to do is cloud hosting. What I'm very against is I don't want to run your infrastructure. Everything that runs in your app is offline and if you want infrastructure, and we are now at the point where we need to add some infrastructure for some features, we're actually going to have you put in your Cloudflare API key, grant us limited access to create Cloudflare workers to do certain things on your Cloudflare instance and then you do this under your own Cloudflare account and Despia is just a nicer interface for doing it that's no-code. So we don't have to manage hosting, because then if something is down, I can always blame it on Cloudflare because it's not our fault. Where if I would be using Cloudflare on my end and something is down, it's my fault. But if we're the interface and they directly host with Cloudflare, and Cloudflare is once again down because they love being down, uh, then it's Cloudflare's fault, not mine, you know. And I don't have to do support for hosting because, you know, to be honest, that recurring revenue from hosting, I'll rather sell them an editor that doesn't cost me anything to maintain.

Yeah, well, that's a good point too.

You know, I was actually going to describe— hosting is a very unthankful business. I have to— I have to actually— I have the assumption that WeWeb doesn't want people to use their hosting. They just want people to self-host because it's way less support and way less troubles for them. They just want people buying AI credits and self-hosting. That's my, my assumption of it. I could be wrong though.

Yeah, I, I don't know how they feel in their heart of hearts, but certainly their pricing does seem to point towards— they're no longer quite as discouraging people going and doing self-hosting as they used to be. The— and I think that's because they understand their, their role is build tool, right? That makes them look more like a lovable, uh, more like, um, you know, these others. The, the— but the— because really what you're paying for are the AI credits on the build side. But I was actually going to say that I look at Despia and I would have not characterized Despia as a build tool. Now certainly I take your point, it's not like long-term infrastructure, you're not running— but you are— but, but you become part of a pipeline, like you're a little bit more of a DevOps tool. Right. Yeah. It's like, yeah, they're, they're, they're expressing their—

I'm using iOS language. Build tool for me means you, you rent from us a Mac computer. That is for me because it builds the application. So that is for me the definition of a build tool. For a no-code person, the build tool is the editor, the drag and drop thing, right?

So, right.

Yeah, we are a CI/CD tool that goes into DevOps. So actually Surprised, probably surprised lots of people. DeskSpace is in the business of renting out Mac computers.

It's the only business we're in. Kind of like at the end of the day, that's what, you know, that's what you're doing, that's what CodeMagic's doing, etc. That's what essentially Expo EAS is doing. The, um, what I was thinking about there was that, you know, you— I hadn't really thought about how you are, you know, interestingly as part of the supply chain, but it's not like there's build and then there's infra. There is build, deploy, infra. Obviously the infra, when we talk about the infra for an app that's running through Despia, is going to be a combination of the Apple App Store infrastructure ecosystem and wherever you're hosting the underlying web app that the Despia is.

There's some things that we're adding. For example, what is if your app is offline? Where we have actually a very cool feature where you, if you have a URL that hosts your WeWeb application, and this only works for WeWeb right now, but if you have a URL that always stays the same, you can add that into the Despi editor. It will be part of the native binary, and the native binary using background processing will always, every 12 hours or so, fetch a copy of your offline application through the WeWeb source, and then once the device is offline, it'll switch to that version that is cached on the device. Of course, you need infrastructure because when WeWeb finishes the build, now this link that WeWeb gets you needs to be hosted on a path that always stays the same, that's on the cloud, so the app can fetch it when it is online, right? So this is kind of like where we need to work with Cloudflare Workers and we need to deploy a little bit of a cloud function for people. So instead of us hosting that for you, we'll either tell you and show you how to set up a template to do this yourself, or we'll build an interface in the editor that'll let you drop in your AWS or your Cloudflare or your Bunny CDN API key. So you have a visual interface to set it up without code, but we don't have to manage the infrastructure because I'll rather have the person pay 30 cents per million requests than having to pay us $5 per million requests. Because if we sell something, we have to make a profit under it. Otherwise we're charity. So I'd rather not sell it at all.

And then at the same time, right, you are, you know, you're not exposed to— they can build with whatever and it was in your— and it will work with your stuff. The, uh, they can, you know, uh, host it wherever and it works with your stuff.

We will probably have it locked in on one vendor that, you know, we think is the best working with Despia. But yeah, if you want to write the code yourself, you can set those things up. Or maybe we'll have like a code generation tool where you can configure your routing and then we'll say, hey, you know, paste this code in your Cloudflare worker.

Here's a tutorial. Interesting, interesting. But, but actually, when we were, you know, talking about build tools, sort of the cycle of this and like the, you know, you have some new stuff some news on the— and some discoveries along the way that you've had with your work in Despia.

Yes. So we have been really pushing WeWeb with Despia because I was looking through the databases, the database in Despia, and most Despia projects that actually upgraded were already on WeWeb. So, but like for the past few weeks and months, WeWeb users actually had a very terrible experience using Despia because those integrations were native. You need to write JavaScript. WeWeb AI hallucinates with it because it thinks there is an npm package that it has to install called Despia, or it confuses it with some random collectible cards, which doesn't really help. So, uh, you know, we built a native plugin that integrates through— that you can install through the WeWeb Marketplace that runs directly inside of your WeWeb app. So you can you know, build native mobile apps and set up all the native functionalities directly inside of WeWeb through that plugin. And it works fantastically. It's a first-class experience. And since we did this, sales have been up 400% of Despia because people now can actually ship their application. Because one observation that I did when we were focusing on Nordkraft is there were many people who signed up for Despia but they didn't finish their projects. They, they sometimes even purchased a license. But then if I look at the database, their project now goes to an empty 404 page. And I think that's very sad. Even though we're making money off it, I thought that was sad that people started building their applications, they heard about Nordcraft, that it was so nicely integrated with Despia. And then they just couldn't, you know, maneuver around it. Where WeWeb is ultimately easier because it is a code generation tool, not its proprietary runtime. AI can help you better in generating code for it, because it doesn't have to learn a language that only 3 people in the the world understand. So when we shifted to being, you know, focused on WeWeb right now for the integrations, for all that, within this one week of having the integration launched, three native mobile apps have been already shipped to the App Store with this new WeWeb plugin inside of it in that week, and they already have been approved by the App Store, which is pretty darn cool. Like, in one week, three apps Approved by the App Store less than a week. I think that is a great achievement. And we had many, many, many dozens of people sign up and upgrade their license and they are already on testbed. I think we already have 20 people now who are testing on testbed in that one week, you know, in Android beta. And that's really cool because before that, people weren't able to ship that fast with Despia. They were stuck at making their componentization work accordingly in NordGraph. So we saw that focusing on WeWeb allows people to ship their applications faster. So we have more success stories in less time for Despia, where in Nordkraft, it used to take weeks and months for a user to sign up to actually upgrade and another few months for them to actually make some progress, get it on TestFlight, ship it to the App Store. So now, within that one week, we've been making more progress, more movement than the Nordkraft user base that we used to focus on has accomplished in the whole past 6 months. And before that, 90%, 80% of the users that purchased Despia were already WeWeb users. So most of the revenue was already coming from the WeWeb community. It just was terribly integrated. So we wanted to do what makes most customers of Despia happy, and that was building a native WeWeb integration. And we're actually working on a full template that is a template that you can start right now and you can customize, add your features to it, and it allows you to set up a native mobile application in a weekend, publish it on Sunday, and have it approved by next Friday probably on the App Store. So it's like a full setup template with Google Auth, native OAuth, Login with Apple, haptic feedback, authentication-gated pages, in-app purchases, RevenueCat, web payments using Polar to completely tax compliant for the US App Store, um, Face ID, passkeys, everything, all the native functionalities. It'll give you a great way to like just import the Xano snippet, import the WeWeb template, have the whole shell already set up, and now you just do the thing that makes it unique.

That's pretty cool. So, uh, can I ask, like, in terms of your, uh, this, uh, the newer WeWeb integration Is this, you know, do you have— was this in concert with anybody over there? This is sort of you hacking your own—

I mean, I remember I did ask Raphael, I think a few months ago, for permission. I said, anyway, we can make a plugin for Despin. He says, yeah, just, you know, put it on the marketplace. And then I put it on the marketplace last weekend and it was approved by Monday. They approved it immediately without any revision requests. So it's now as a library. So you have to import it as a component, but that makes more sense because then Despia only runs when the DOM is initialized, which is a big problem in WeWeb because it's a different lifecycle to what non-View developers are used to. Because with Vue, you have to all worry about the lifecycle and when does it run because it's client-side. So it actually helps with a lot of things, making it as a component, not as a plugin.

Interesting, interesting.

All right, so I'm very excited about WeWeb. I'm right now— I'm like, I got 3 WeWeb projects open. I have been already finding my way around WeWeb, and I can actually now design stuff that looks pretty darn good in WeWeb, which was always like my critique. I was like, I can't do nice stuff in WeWeb. I just understood that it was a skill issue, because if you put in the time, you can do pretty nice stuff. And do you see my interface here?

I can see it very tiny.

Does that— let's see. Oh yeah, I got to make that bigger. That's my mobile app so far. I'm working on that. That's the login screen. You got login with Google, login with Apple. I'm going to make the dashboard. There is a native drawer coming looking like the iOS one being touch responsive. Think of it like Framework 7, Yeah, yeah, it'll be like a little bit like Framework 7 plus like that Ionic starter pack plus better auth for WeWeb and Despia and Xano. So I'll give you like this whole boilerplate to set stuff up so you can ship an application in a weekend, so you can buy a paid Despia license in your first 3 days of being signed up.

That, and I think Lucas, uh, took the, took the word right out of my mouth.

Nice, nice. Yeah, but you know, I, I think I, I do, I want to see people make progress with Despia, so I want to make sure that we have the integrations, integrations with the tools that help people make the most progress the fastest, because I want to see lots of Despia apps on the App Store. Being very successful. That's what keeps me going.

I think that's a great North Star. I have told the story before of talking to other companies that are in the, hey, we help people make mobile apps. I ask how many apps using your platform are in the App Store today? Then I get a mumble, mumble, mumble answer. Not for nothing, but here we are 6 months in, What's that number for Despia?

It's about 60, about 60. But if you have to, if you look at that again, a week ago it was 50 about. So like we're now getting, uh, 2 weeks ago. So we're now getting like before that, even before we launched the WeWeb plugin, because there was the announcement that something is coming, people signed up for it. So like even like people were shipping apps and the past WeWeb apps before there as well. I would say of those 50 to 60, I would say 48 are WeWeb. Wow. Yeah. Like we got one or two apps on the app store that are built in Nordcraft. Three, actually. One, he's almost there, but I think he's on Android. Only. So like we got like 3 Nordcraft apps, most of it around 50 WeWeb apps. And then this week we got 3 approved on the App Store again. So we're getting there where we have those numbers that we can show and those case studies, right? Actually, one of the first apps that was approved on the App Store built on Despia was a WeWeb app. I was one of the first users. I never wanted to look into WeWeb because I just gave it a try for a day and I was like, "I don't like that thing." But that was right after Wist. To be honest, I think I didn't give it a fair shot. But now looking back at it again, I actually— I don't want to admit it, but I think I may have been wrong about WeWeb. I think I'm actually falling in love with that tool. It's a very nice tool. You can just get Vue components in there. You got access to the npm package. I can run npm packages. I'm not limited to putting stuff on a CDN. With Nordcraft and Wiz, you got to use jQuery plugins that are older than I am. Where now with WeWeb, I can use the new hot stuff out there on the NPM ecosystem that doesn't use a CDN anymore, which is pretty nice. Because it just builds the component, right? Like, they have this cloud building that builds this component, compiles it for you, which is pretty darn cool.

[Speaker:JASON] Yeah. I'm— yes, they've done some really good work. I've enjoyed WeWeb since— well, actually, frankly, before I started StateChange, when I got to first meet those folks and then, um, you know, get to see what amazing things people have built. Uh, that's what actually— that for me is what I mean, bringing— but the beginning of our conversation, that is what, uh, you know, gets me going about SageMaker. So you can see all this remarkable stuff people have built. Like, they're, they're doing this— hey, I've got this hard part, this is the place I'm stuck. Like, I can't believe that you did the part that you did. Not because it's— I was surprised they could do it. I'm surprised these things could be done. Just like the, the amazing variety and like the, uh, the way that the WeWeb has allow that unlock on the front end is really just a credit to that team.

But also, I do have to say, my number one goal with Despay is to get as much apps on the App Store, as much beautiful apps, as much high-quality apps. And we have to put our attention on the tool and build integrations for that that allows us to get those results, because those results are only what creates trust. There is some trust that I create by just being around there and jumping in front of the camera and saying, hey, this is great. But The ultimate trust that we're getting as a company is based on results. Now, we need to get those results and I found it's easier for us to get those results so we can show it to Nauticraft users, we can show it to lovable users or whatever, we can show it to React users and tell them you can do this with Despian and it actually doesn't feel like you remember your last WebView, how it felt. It's cool. All right, so for us using WeWeb, because we get more good case studies, which shows that it's a great tool for building applications quickly and with high quality.

Um, okay, so you have all these for this product development, uh, in the course of the last bit. You've got the, like you say, the big uptick in sales, what have you. Um, you know, we're, we're going to do this again in a week. What's going to be true when we talk next week that's not true today?

I think next week there is a good chance that we're going to have NFC set up in Despia where you can read and write NFC tags in with your WeWeb app. Usually WeWeb apps can read and write NFC tags with the native sensors. So now going to integrate the native sensors. I remember last time you said you like playing with the native sensors of mobile devices. We may get you back into WeWeb then.

Yes, the native stuff really just is where the greatest fun is. I really like that bit. I was thinking about that relative to, think about OpenRouter and how there are a bunch of people who like OpenRouter because you can swap among models.

You will love this. We will have access on the desktop mobile app on compatible devices to use the GPU to run AI inferencing on the device.

Oh, that's really cool.

Does it make the app hot? Well, in our internal beta, yes, on certain devices. We're still working out the kinks, but we're working on supporting local AI. So this will be the foundational AI models that can run on the device. Also, we're going to create our own database that will not be hosted. It will be called DespiaBase and it will be SQLite. So you can access a comparison to SQLite. It's not SQLite, we can't use brand names, but it is inspired by SQLite. So you can use something that is inspired by SQLite natively in your web application. And I remember there was Scott from StateChange and he was like using IndexedDB, and Fred had that issue, and everybody has the issue How do I get my web PWA to work offline? We don't have to worry about that anymore. There will be a plugin for Despia Base that lets you read and write and query any value on this local database that your application will include. Your whole application will be able to work offline because you'll be able to put a Despia webhook on your web build and that will be the offline version of the app. The online version still has updates over the air, and then the offline version will sync itself every 12 hours with this, you know, depending on how we'll host that. We don't want to host for you, but we'll see how we set that up. You may want to host it yourself. You may, you may build an interface for it, but you'll be able to have fully offline working WeWeb apps, and you can cache all your WeWeb collections inside of the Despia base. You can even store videos in there to allow offline playback. Of course, you'll end up storing them as, uh, you know, base64, but it can pretty much handle it because you're using native hardware acceleration. Yeah, so yeah, this is, this is groundbreaking for WeWeb users because PWAs usually work terribly offline because you got to cache stuff and then, you know, like, then the app isn't real time anymore and all that kind of stuff. Where we have two versions, we take a copy of your app from the code export, that's the it's the offline version and then we take the live version from your URL and they switch. And then you can set the state of them in a database. So you can have a database that's like AppState and then if they are on the home feed and then it goes offline, you just open the offline app and then look in the database, say, oh, start at the home feed. So it's like a seamless experience. It just blends in. They won't even know they're offline. And then you can store the request they're doing in the Despia offline database. And then when they're connected, In the online app, on load on the online app, when it reloads, you will look into the database. Are there any tasks scheduled? And then you send that to Xano and work with it.

That's super cool, man.

So many months to figure out.

We're looking for that for, for next week. We're going to see what progress we can make on our work together.

Uh, you know, for that, I think, um anybody is really serious about building an application, they gotta be in Stay Changed. But I also think they gotta be in No-Code Pro-Code if they are using WeWeb, if they are using Norecraft as their front-end tool. And with this new $199 a month plan, you can get access to both. You can get up to 10 hours of office hours a week. You essentially have two not-official CTOs right at your side, uh, looking over your shoulder and assisting you to make sure that you're going into the right direction.

Yep, I, I, I, I think this is a— I, I, I, I do like the idea of the offer. I, I, if, if even just one person does it, I'll be just so excited about like the opportunity to help them go just insanely fast. We talk about like how some of the stuff that just can't really change, but the thing is you can go faster. It's just not— you can't go faster just with a tool, right? You go faster by having this, that little bit of input to the skill sets you don't necessarily have at the beginning, but you do have at the end, and, and using the best tools for the job. And then just, oh man, the sky's the limit.

And now this also extends state change to be Nordcraft compatible. Look at that, for all the— because Ray, you've been like avoiding that like a hot potato, you know. Every time I bring up Nordcraft, you're like, I'm playing with web components. I'm like, what happened to your efforts in learning Nordcraft?

Um, yeah, well, actually, you know, that, that question of like what people should be spending their time learning, I think that is a great topic for us to pick up when we do office hours next time.

Yes, we'll be prepared better and we don't have to reschedule anymore because now it's every Monday.

Yeah, I'll see, see you again, same bad time, same bad channel, man.