#22 OpenAI Dev Day, Major Code Tooling Updates, Webflow and Framer Updates

How does that look? I've clicked the 4K button. Can you see me better? I mean, this is a 1080p c Well, it's not a 1080p camera, but like I think I can only push 1080p through the Anyway, whatever. Point is, we're live. Kabaza is just finding out his earphones because he keeps ruining our audio. You're trying to raw dog our podcast here, man, without earphones. You know what? What's funny? Well, we are now live on Instagram as well. Uh, you saved these. If if it wasn't for you mentioning it, I would throw them into the washing machine with the rest of the Oh my god. What? They were in your dirty laundry or something? Yes, they were in my dirty laundry.

Hello. Hi, Tobias. How are you doing, guys? So yeah. So now I'm going to wear them so we will have like better audio in celebration of that saving. I mean they're not that expensive. Are they Apple ones? They're not that expensive to replace, are they? Yeah, these are old ones. I was thinking about those the other day. Like it's so nice just to have something that doesn't need batteries, doesn't need charging. You could just plug it in and go. Like I miss those days a little bit. I have dungles, so I hate them. I I want my uh AirPod Pros, but the case is broken. The case is broken, and I don't want to buy a new case because I want to buy the new new new one.

Sure. Sure. Yeah. Yeah. Fun fact, so when I replaced my AirPods with the USBC ones, um I made more money selling the case. and each AirPod separately than I would have made just selling the whole thing. Oh, because people do people still buy Yeah. the AirPods, you know, especially when they lose them. It's like what, you know, some people are icked out by it and I don't I don't blame them, but some people like, yeah, whatever. Just stick a new That's why that's one of the reasons why I don't buy and it's the case like the AirPods are fine. The case is broken, but I'm still I don't like the idea. So, I don't like sharing them.

Yeah, makes sense. Makes sense. Shout out to my mate Smid who will always like buy broken and then he'll kind of fix it. So I bet I bet he could open that up or just you know I want to but they are pretty like bad. So these are like as bad as that it gets um when it comes to like repairability especially the new ones. They have I believe something like zero or like one from iix it. They have like one of the worst ratings because they are not fixable at all. But I do fix things, by the way. I do like I I genuinely enjoy fixing broken stuff. As a man, it's our duty to fix things.

Yeah. I mean, I have a tons of tools and I like like look what I have on my desk. I have a bunch of like I have so many of these batteries. Like I have What are these batteries? They look massive. These are not normal batteries. What are they? These are double A. They must be like single A. Well, uh, these are I don't know what Well, I actually know what they are called. They they have it written on them. They are 18, uh, 650 1850 batteries. So, it's just a naming. Um, and these are from my power bank. So I I had a power slip there. Yeah, I got a power I got a smart bong that I can uh link up to my phone.

So it it but it was so big that I couldn't like use it properly and I had a better one. Uh so what I did, I opened it up, I took these out and I can actually use them. I use them uh in my torch. So I have a bunch of like I I I am a man with like a billion tools, so I I can use them Yeah. to explode. Yeah. So, very good. How have you been doing? Been doing good. I I would have uh I would have let you down this week, I'm afraid. Um the last two days or the last two weeks, right? You've actually No, no, no, maybe not the last two weeks, but last week's for sure.

Then previously you've you you always raised the question like do you watch like AI content or do you like like and you're like you actively skip it, right? I have been balls deep in listening to AI music the last few days. Like it it's not it's not just like randomly generated AI music, but do do you like Blink182? I have no idea. Okay, fine. Anyway, they were like one of my favorite bands of all time. Yeah. And and I find actually No, no, we we'll go right back to the beginning. I get sent Do you know corn at all? No. Like, okay. No, that's disturbed, didn't it? I don't know. Anyway, um my mate sends me a soul version of a corn song.

Now, corn on like heavy metal, new metal, new metal, something like this, whatever. And the soul version is just so good. Um, maybe I've got it in my history here. Here we go. Copyright free, baby. We don't have to do this. Is it AI generated? AI generated, you know, just chilling, but the original is like, you know, crazy. So, it's like you hear the lyrics. It's really cool. And then I I discovered this guy called Professor. So shout out Professor. Um but then I just get started give get given all of these AI covers because they've got so much better now. Like I remember I was sent Slipnot, you know, must know Slipnot. Yeah. Yeah.

Yeah. So I got sent a Slipnot or I found a Slipnot version. It was like it was good but you could tell it had an AI voice on it and stuff like this. Anyway, I'm on this rabbit hole and then I get recommended. I'm slightly drunk last night and I get home and I I just get recommended on my feed um Blink182's Adam song. Um like a um I want to show you the I mean maybe you can see I don't know but like the covers are like this. Can you see that? Yeah. With black guy afro and stuff like this. And I'm just like jamming out to this music last night. Like m maybe you know the biff.

It kind of plays around with the pitch at the moment. And I'm just like And this is AI. And this is AI. And then you get to the chorus and I'm there's I've just got chills running down my spine cuz it's so good. That's actually pretty good. It's so good. So you're listening to AI music because this song is a very um it's a dark song, right? He's talking about suicide and stuff like this and but the the juxosition of that song being so like happy and like um optimistic. It's like it's just like really cool. Um but it's such it's so good. like I just listen to it thinking how it's almost better in some ways, but it's a very it's a it's a minor song and that's obviously a major key uh version of it and like yeah, I'm listening to it and I'm trying to work out how they do it because you can't just like whack a song and say make this.

I I do think there's a level of craftsmanship that has to go into that. Like maybe they take the lyrics and then they plug in a melody and they do the individual tracks that then bring Do you know what I mean? Like it can't just be like a one it can't be just like a oneshot wonder. I have no idea if you know if anyone knows how they do all this. I don't know. But saying all that like this music first of all I remember watching a video actually. First of all this music couldn't exist without the real version. Like yeah it it's it's novel to me because it's like Blink 182 will would never produce that version like would never make that music.

So there's a comedy in that and there's a just a um like a what do you even call it like a parallel universe type of aspect to it. So the original needs to exist for that to take place. And again we get into the discussion of like hating AI art and stuff like that. But at the same time I don't know there's just I feel like there's a place for that that doesn't outwardly just uh it doesn't critique anyone. You still get people in the comments obviously. Once I realized this was AI, I shut it down. It's like I saw one comment in the corn one where someone was like, "I enjoyed this until I found out it was AI."

And it's like, if you enjoy it, you enjoy it. Like, what? Like, I don't know. Anyway, but anyway, I thought I'd tell you that I've literally been actively going on to some of these songs and um of the rock genre and and listening to these soul jazz versions and I'm sorry to let you down, but No, I Well, I have listened to AI music in the past. Oh, okay. The truth comes out. Yeah. Yeah, I have. But it's when when somebody makes it and it's intentional and it's something behind it, why not? Like I have no issues whatever with that because in that AI is just a tool used by a creative to to do whatever they want to do maybe better, faster, whatever.

Um I don't have an issue with that. So, I've I've listened to music, AI music, but I'm I'm laughing because I've been sent a lot of uh I've been sent a lot of uh Sora made videos, like AI generated videos, like my DMs are full with my friends sending me those videos and they are mostly well almost all of them are memes essentially like Albert Einstein doing something funny. It's kind of like it has become a new meme generator. We can talk about it later in the show, but we've been promised AGI and we have a glorified meme generator. Now, if if you've read um AI, what's it called? Iroot. No, no, no. Yeah, I one of it's one of the first sci-fi books ever.

He starts by saying, "We're going to use robots to set clocks and check the weather and stuff like that." And this is exactly, maybe we won't ruin it right now, but this is kind of heading in the direction of the conversation where I was saying before the call, why do I keep calling it a call? Before the show, um, I was saying that AI is in this bubble like we're trying to do so much, but actually it will just it will just settle into just having fun with it. You know, we think it's going to change the world. We think we're going to turn us into dumb people and this and that and it's like just Have you seen Ideaudiocracy?

Um that's the film. No, it's uh Idiocracy. Idiocracy. Yes. Yeah. Yeah. That's I've seen it many many years ago. I couldn't I couldn't reel off, but I know the idea, the concept. It's it's kind of like Walle, but real. So, we get dumber and dumber because like Yeah. Uh it's it's quite funny. So, if anybody wants to watch it, um it's it's not like the best movie, but the idea is pretty good. It's like stoner comb bomb coming out. It's it's kind of like stoner. Uh I think there's a name for it, but it's in the same sort of realm as like scary movie and um what's those two smoker dudes? Um uh something Chichin Chong is it?

I don't know. I don't know. Anyway Anyway, it's it's funny. It's good. Yeah, that's my uh that's my anything to raise before we jumping into the into the No, we will talk about the meme generator uh and Sora in a second. So, well, once we get into it. So, maybe you send me a few like good music. Um like I think you need to know the original. So, like you tell me what music you like. I think that's where my enjoyment comes out. Well, to be honest, that's a kickass song, but you know, maybe I should close my window. Is that Can you Yeah, I can hear it. Yes. I'm gonna go close my window.

Just do some magic again. Yeah. To what magic? So, a magic trick. So, I'm going to show you. No, I'm kidding. I have my cards always here with me. Uh I'm I'm practicing. I'm pretty good in person, but I have never done it uh on camera. So, let me see like if people are chatting in on Instagram. No comments. So, a lot of people joining and leaving. Uh, guys, when you join, there's a rule. You have to stay. You cannot leave again. They just come in and think, "Oh, AI music. this." Yeah. Yeah. Essentially, so you have to stay, guys. Yeah. So the background, you know, we have some people on Instagram, so we we can we can start we can start.

We we have quite a lot to talk about. And after the show, I have to I don't have to leave too early, but I have to leave pretty much on time. Same. Actually, I've got a party to go to. Yeah, I Oh, a party. Um Well, I have to do shopping and after shopping, I'll go shoot my friends uh in laser. dark real quick, didn't it? Uh, so my siblings and my friends, they have apparently they have a scheduled laser tag and I just found out like an hour or two later uh before that. Um, yeah, we have we have some laser tag to do. Hey, you just found out or you completely forgot? No, I well they we have a group with the boys and I didn't know if they have a plan for tonight and apparently they do have a plan.

Uh and the plan is that I I will shoot them. Cool. There you go. Laserw. Okay. Cool. Let's do the intro and uh yeah and get this over and done with. Yeah. Okay. All right. So, you ready? I'm ready. So, this week we had Open AI Dev Day, Chrome MCP, uh a new framework pricing, and a few things about Webflow. All of this in this week's episode. This is Command AI. I'm Karza. And I am Sam, and this is a week in AI design and dev. Cool. So, should we start with Chrome MCP? Let's Do you know what? That's a fabulous idea. That's a wonderful idea. So, this is really cool. Actually, this is this is So, can you like explain um your um intro?

Let me do the then we can chat. Yeah. Do the intro like I'm two. Explain to me like I'm two. Forget Playright. Forget Cyprus. Forget all of these automation tools. Chrome have just released the MCP server to take care of all of your diagnostic needs. So what does that mean? Does that mean that we can use a c can can we chat to AI to use Chrome? You actually mentioned this. This is why I kind of wanted to bring it up because you asked whether you can inspect the Chrome dev tools and and look at the code and do this and do that. um this is what enables you to do. So MCPs are basically an API into a tool and this will become very relevant later on when we talk about uh dev day.

Um they've basically given you an API that you can modify and do things and look at things in the Chrome browser whether that's diagnostics when it comes to like running um have you ever done Lighthouse and things like that? So you can look at that sort of stuff. You can look at the the code. You can look at what's on the actual screen itself. So they've given you an API there and the MCP layer, you know. Does that make sense to you? Let's break this down. Does that make sense to you? Yes. But does this happen on Yes. But does this happen on my machine or is it like uh on a virtual server? Like can I this is on your this is on your machine.

So your your web page you're currently looking at right now inside of Chrome has now an API wrapped around it. ah okay that you can that you can now interact with using a tool that has MCP functionality so clawed desktop um warp uh I don't know what what do people use for MCP um so you can say you know uh Chrome MCP server what's on the what's what browser tab are you on something like that or what's the current Lighthouse score of this page something like that super cool yes that is Super. So I can theoretically can I like build a project with web flow and then use cursor to know what I'm like building.

I don't know if you can take action on those things. Um I don't know. I literally don't know. But yeah, you can look what can you use it for? Verify code changes in real time. So what browser uh verify in the browser that your change worked as expected. So this is obviously if you're developing, right? if you're writing some code and you just want to check that it actually appeared on the front end. So this is where we get into you know um testing and things like that. So maybe no more pasting of my console log into cursor. So cursor can just read it. Cursor can just read it. Exactly. A few images aren't loading.

What's happening? So it could and then because it can read the console, it can look at those console logs and and do all that. Where is the network tab? A lot of times I had issues with exactly navigating the network tab and understanding what is not loading. Why is that uh not working? So look, it can take action. Why does submitting the form fail after entering an email address? So this is obviously has to submit that form to be able to give you what the errors are in the console. So this is this is really really good. Um and performance audits. Look, this is what I was saying about the the um Lighthouse stuff as well.

Live styling and layout issues, you know. So, we can see the code, you can see the CSS. It can be really really handy. Um and it's just a simple case of adding that into your into cursor. That's all you got to do. How does it come to play uh with something like uh you know these agentic stuff that Chrome is trying to do and of course something like DIA browser because with this does that mean that the AI layer that we will have will have like more control over whatever uh page you are on. Well, this is obviously geared towards devs and designers, right? I don't think that the average person is going to interact with Cro through MCP.

Nothing stopping them doing it, of course, if they want to, but like this is going to empower us as we code and we work to have a have a workflow and we we can get into this have a workflow where we're making changes in our code and we're just verifying that things are happening over in Chrome properly. So that I don't think it has any well it doesn't demonstrate any um compatibility with the AI layer that GIA has or or comment has or or whatever. There's no MCP or there's no API to interact with that sort of stuff. We're just interested in the code that actually gets rendered on the page and the logs in the console and the network traffic and stuff like that.

So it's really the dev development sorry side that that it empowers. So it's really cool. This is really interesting like Yeah. Uh I really love the idea and this is going to open so many doors. Yeah. No, for sure. So I've been really going hard in trying to sort of formalize an AI coding workflow and one of the big things for me has been MCP servers. And so I now have I've got an MCP server that goes out and searches because generic webs there are MCP servers that actually search specifically GitHub and coding related stuff. It's like it's like Google for coders only ever returns code is results and things like that. So that can obviously refine you know when claude code goes and does a web search it's going to get a bunch of results from nothing to do with development.

So, I've got Exa, which is my MCP server for doing that. I've got REF, which is up-to-date documentation. And so, with that, I've then got rules and sub aents that pick up on whenever I do this, use that MCP. Okay. And one thing, and I released a video to my patreons uh yesterday, which is a real indepth look at how I'm using claw code and doing the rest of it. And I I finalize it by saying one thing I really want to get better at, which is where Chrome MCP comes in, is the actual looking at the rendering and bringing that into the life cycle. So I need I I would write a rule to say something along the lines of verify, you know, this and that or just just getting it in getting it's more habitual.

You got to get into a habit. I think you've got to remember the MCP servers that you have and say, "Oh, do check that." You know, I bought Oh, go on. Do Do you know Unlouse? No, no, I don't know that. So, you mentioned lighthouse. So, we have unlouse. It's a funny name. It's literally Google Lighthouse, but for your entire site. So, it um instead of like just uh lighousing one page, it takes your entire website like all of your pages and it tests everything. So, it's really cool to basically just do this with one command for your entire website. Yeah, I was thinking if this can do that. I don't know if I like that idea.

I don't know if like, you know, we talk about context. You you start to try and broad like I can I can see the appeal of like testing the whole site, but page by page. Like who wants to be testing all of this stuff all at once, you know? I've got to unshare. Yeah. Um I want to know how to bring my page down without removing it from Yeah. How can I like change? Because if I remove it, boring stuff now, but in the edit, this will if I remove it, it'll just send me the link. Send me the link. I I didn't want to sh you know, I want to share the screen and then change it to mine, but I it seems like I can't.

Should I You can't swap it. So, I can't I can't share it. It's not me. Okay. So, it's okay. It doesn't matter. Uh I want to show Oh, actually, wait. Wait. Uh can you like I think I can do it. Yeah, I know. Okay. How did you do that? How did you do that? Press remove on on the boxes that we have below. Oh, I knew that. I knew that's what I was trying to avoid. But all good. I I thought if I press remove it cancels the screen share but it's just remove okay so yeah know it used to I think I used to so that's good to know yeah so this is on lighthouse and it gives you this view which is really cool look you wouldn't run this all the time you would do this like once every so often this is too much to do after every change I like you give yourself a break I think a little bit and like like yeah I I don't know why I thought of this because when I thought of an MCP that can do you know at something of with a page that I give it to maybe it could do it with multiple ones and yeah because I mean yeah to bring it to talk about the AI side of things and improving dev workflows and stuff like that it's becoming more and more apparent to me that you you cannot you you get much better results if you're very isolated and focused on one specific area.

And I think that goes in in this case as well. Work on the performance on a on a page by page basis. Like what what good does it I mean the one thing I I can see that's quite good about that is that you can test the whole site and then if something is coming up red as in really bad, there's your priority. That's all I can see that that's really good for. But either way is, you know, whatever. Um, yeah, cool. But yeah, so MCP servers are becoming a big part of my workflow. The only thing to to say about it is that maybe for not your um your case, but MCP servers eat up context.

So I noticed I had a bunch of MCP servers like enabled for my website, you know, as I'm coding and I would run out of context very very quickly because it's eating up loads of context. And you can check your context and claw code. Um, uh, you can get like a nice diagram, uh, and you see that MCP adds up. So, I removed a bunch because I'm like, well, I don't really do that. I don't really use that very regularly through my coding workflow. Like, I'll just I might as well just do that outside or I'll do that once a day or something like that. I just don't need it on a I don't need it like within arms reach.

So, I just turn off that MCP to save myself a bit of context. But this does like something that would work its way into my 100% like I'd need it all the time MCP server. I just need to build that workflow. So yeah. Um I I got a question on that. Uh we might need a AI engineer to answer it. I'm not sure. But do we like get um like do we use context exponentially? So as we chat with the AI system, do we use the context exponentially? Meaning that from what I understand, every time you chat to the AI, every word all the way back to the first letter is given as a kind of like the the seed back to the machine so it can produce the next word.

So does that mean if you give it a huge context and then you continue chatting, you're eating it out exponentially? So I'm glad you asked that. I'm guessing you haven't watched the fundamentals on the cursor we was talking about last week. No, but I watched a similar video that's why probably. Yeah, but I did explain this last week, right? And this is it a practical use case. No, it doesn't because it's stateless. Every individual message you send is uh completely wiped. it is exponential in that every every um uh you know every request that you send just adds up and adds up and adds up like the memory but it doesn't it doesn't repeat itself because it's completely stateless you know so it just it just adds a um a foundation to the to your prompting already you know yeah and thank you for the reminder I completely forgot no uh I I watched I watched a few videos from the channel three blue one brown.

Okay. You don't you don't know the channel? No. Well, it's a it's about mathematics. It's it's the best channel on YouTube when it comes to anything mathematics. Three blue, one brown. Amazing channel. Uh very deep, very niche on mathematics. But um it's like number pie. It's also another mathematics channel. Uh, and in that channel he talks about also AI, how AI works because it's essentially just mathematics. And he he does break it down in some videos. He breaks it down to very simple terms. And in in some other videos he goes from simple into depth and like actually like how everything is being calculated and how how the 3D vector mapping works for language. It it's it's amazing.

So 100% um recommended. Recommend. Cool. Cool. All right, let's talk about Dev Day because this is pretty much where is it? I need to add it, don't I? This is This is I mean it's huge. I'll do the intro and then we can talk about it. Open just dropped a bomb of new products and some of them have got me extremely excited. So OpenAI Dev Day is a is a conference as you can see here. They released apps and GBT which is the thing that I'm very excited about. Um agent kit saw 2 API uh updates to codeex making it you know publicly available uh generally available they call it GBT5 pro in the API as well.

GBT realtime mini uh GPT image one mini um just a bombardment of new uh products released. As I say, the GPT, the apps inside of GPT is the one I'm particularly excited about. Um, did you watch it? Did you follow along with it or anything? No, but I watched the videos like after I didn't watch the conference itself, but I watched the like reviews and stuff. Yeah. So, apps apps and GPT is pretty cool. And it's it goes back to what we're talking about about MCP. I think the fundamental aspect of this is is MCP. I think it's a web view that has an MCP attached to it and then it just you've just formatted it to to sort of sit nicely within the the chat.

But this is basically, you know, you can see here Spotify use Spotify for this answer. You're this is how you interact with MCPs. You mention the MCP or you say something to do with the MCP that it recognizes and it'll be like boom. You want to you want to chat about is it like a tool call? Well, yeah. Yeah. Exactly. No, it's exactly what a tool call is. But but these are built purposely for or you know I think you can again I need to look into the technicalities of this but I think it's basically just um uh media queries you know or some form of um you know recognizing that it's in chat GVT I don't know I'll look at the SDK but I think this is all web technologies I don't think this is anything crazy I think it's just typescript and you can build you build your app or you you format your app in such a way where it just looks nice inside a jackp PT and you reveal a MCP server to the chat and it can work with it.

And I really like this idea. I think I I do wonder where this is going to go because this are people going to use apps inside of chat GPT? I don't know. But this may this this is this is crazy. This is we used to say like um you know when the AI application uh sorry products like the rabbit and the pin things like that we think well our phone can do it. You know Marcus Brownley just you know he had this pin on him and he asked what car it was it was a Tesla or something and in the time it took for this pin to get back he just snapped it with Google Glass and then he knew what it was on his phone.

So we used to argue that AI, not even AI, sorry, these AI products could be replaced by the phone. This is then saying, well, now everything is in the one app. So yeah, you don't even need to touch any of you don't even any apps on your phone. We probably do for this to actually work. I don't know whether this is done through the web or whether it's actually interfaces across to the app. I probably would assume it's the website because then that gets past Apple's nonsense. Do you know what I mean? Yeah. Um which could, you know, you could bypass Apple's app store 20% thing by allowing people to purchase subscriptions through chat GBT.

Maybe Chat GBT will take a cut. I don't know. But this is probably Apple change the policy for this. But I I have some feelings about this. So there are aspects of this that I like and there are aspects of this that I think are not new at all. So talking of what I like and I've been talking about this for quite some time that I don't like a chat interface like I don't like reading that the long answers that chat GPT gives me and I find it fundamentally like un not fundamentally like but I would say I find it most of the time unreliable and not doing what I want in terms of the amount of text that I get from chatp.

Sometimes it's too short, sometimes it's too long, and I find myself specifying how much I want from it. So, all of this to say, I personally don't like this chat that I get. I like talking to my phone and getting a UI. That's where we get into the good stuff. And what I've been saying that we need more UI, not less. We We went from essentially no UI. We had um we we didn't have a UI interface for computers at all, you know. Mhm. We started with that. We started without UI and then we built UI and then we had this glorified UI and now we went back to just chat again but in a more human natural language way.

And I do believe that we want to have some level of UI because we want to see options. We want we we are so visual based uh and yes text is somehow visual but it's not exactly like this visual base that I'm talking about. We want to see like pictures, shapes, um structures and this I like it because it gives me this like UI but I would like to have more control. Another thing about this is that fundamentally nothing about this is new. Like what is this that theoretically a good Siri couldn't do? Theoretically. Yeah. So Siri is trash, right? It it's a good point because a few weeks ago we spoke about um MCP, Apple releasing MCP in their operating system and this is exactly that.

you know, this is Siri being able to control the apps themselves and whatever. Let's let I mean, I don't think Siri it's not going to be any like where do where's the difference? Where would the difference be? So, the difference is like here in Chachi PT we have one app within an operating system that is inside of a phone. So we have the phone and then the operating system and then an app and then an operating system essentially within that app. Yeah. So instead of this I think we we should move towards like having the AI on the operating system level and also on the operating system on our computers maybe also on Chrome because or Chrome or like browser because our browser is kind of like this window to essentially like 90% of what we do daily.

Um so what I think it should happen is to have less chat and more agentic stuff. Yeah. Yeah. And essentially kind of like this glorified idea of an assistant that all of these companies are trying to tackle in different ways. And this to me is just a different way in the examples that they showed. I don't like the the one from um what's it called? Like Spotify. I don't like the Spotify example. I feel like I wouldn't want it to work that way inside of chat. I don't want to go to chat GPT and you know to a new chat and start like but you would not do that. That's not that's not what they're saying here.

You you have Spotify in your app if you just want to open up and play some music of course. But as an example, I went to a concert and I took a photo of the set list and on the piece of paper on the stage sort of thing. I took that I posted that to Chachi BT. I said, "Can you go off and put together a playlist for all these songs, right? Did it unfortunately because of a lack of an app ecosystem or something like that, I had to get a text thing and I had to copy that into whatever blah blah blah." This could be it. It's it's convenience. is bringing it to you into a conversation that you're already having.

Maybe you're trying to travel plan. Maybe you're saying, "I'm going to Albania next month on a road trip. Um, where's the best place to go?" Blah, blah, blah. Okay, let's book a place with Zupla or wherever, you know, it's it's it's the con bringing the convenience to you whilst you're chat already chatting about I see kind of like project it. Yeah, exactly. So, yeah. No, I I don't think that's what they're trying to achieve. However, I do get your point that um Siri because it's it's a it's um a more higher order. I don't know whether that translates. It's more higher order like more controlled theoretically like that. But the long story short, Apple aren't getting their ass into Git.

Yeah. You know, this is like this is ChBT taking the power and being like, look, if you're not going to move, we're going to move. All right. And it would be great to have both options but like whilst you're chatting away you know you use chat GBT like in the middle of a conversation you you might have come to a conclusion or booking some as you say projects and then you can bring your thing I think um but it's a really good point because like I'm excited by it and I was like oh where can I bring what can I do with this where can I bring these apps in and I thought of a app that I'm building my friend and I'm like I think this would be really good use case for that.

But I think your brain has got to understand why people are in chat GPT in the first place, which can be a a billion different reasons as to why. And it's up to you to figure out why people use it. And why is your app relevant within the context of Chachi BT? Because again, like you've just said, why would I go to Chachib to ask for this, that, and the other when I could just go to the app and get that this that and the other? You know, you it's um there'll be an analogy or or whatever somewhere and that you know you some shops are in train stations because that it make a book shop is in a train station because most likely those people are in the train station because they're about to go on a long journey.

Simple marketing. So does your app is is someone in the world of chat GBTE like does it make sense that you're you're you're cross-pollinating those roles you're putting yourself where you it's a marketing thing right you're putting yourself where your users are already and now I'm thinking probably not with my mates app they're probably not in the world of branding um in JBT probably maybe well we do that as well but I feel like Chad GPT, they they have definitely a lot of users and I feel like they're trying to to try a few different things and see which one sticks. They're trying like to see they have 800 million weekly, I believe, users. Crazy.

That's a lot of users. That That's a lot of users. Um but I kind of like feel like it's really difficult for Chad GPD to win. they are in a such a difficult uh position if you look at operating systems and AI uh being on an operating system level because we have Google and they are good with AI as well. So could go anyway. And I think that you've just got to literally, as you say, throw at the wall, see what sticks because they we will find our place. And there'll be a whole number of goes back to that video I said about Dave Shapiro released about the economic, socioeconomic um impact of everything. Something might happen that contextualizes everyone just lives in ChibiT for the rest of their lives.

I don't know. point is they've just got you've just got to experiment. You've just got to release. And on the one hand, I sympathize with the fatigue on AI, all the things. On the other hand, it's like you've got to experiment. So, yeah. Um, but anyway, moving on from that, the agent kit has also got me quite excited though. It's, you know, did you see the agent kit? Do you know what this is? Yes. I I watched a few videos on it. So, yeah. Yeah. So basically na have I got a this is probably the most interesting one by the way. So agent kit just to be clear agent kit is a suite of products right a suite of things.

It's an agent builder is the connector regist registry and the chat kit. So, it's these three things, but I'm I'm assuming you're talking about the agent builder, which is a noode interface for building agents, which looks sweet. And we probably won't cover it because it's boring as but it's the Google Enterprise thing today. They've also released an agent that you can build that operates across Google, which I think is just really much more powerful than this because it's funny you just said that, you know, they're in a difficult position because of Google because Yeah, Google. I'm on Google Workspace. I presume you're on Google Workspace. Yes. You know, we can build agents that work across everything that we're already doing, whereas we've got to integrate this with our But regardless, it's still pretty cool.

This is going to be really cool for people if they can get their head into this world of, you know, automating and building things. And can you please like zoom in just with your trackpad like zoom in? Oh, yeah. Yeah, I know. I got you. Yeah. Because Yeah. Like zoom in more like more even more like look at the UI. It's no no like the details of the UI. It's really sweet. I love their design. It's so It's so clean. So fun. Any AI tool NATN take note. This is so approachable, so human. So I feel like Johnny I've worked on this. I mean, who knows? Maybe he had some input. And it's funny you you talk about N because you you you got some news on N, right?

Yeah. Did you want to share that now or later? We can talk about this. Yeah. Now, just because just because I think we talk about N being a very very great candidate for the FE or maybe are they aren't they? I think this is something they need to take note of. like you've just said, they just need an appro you need you need um n light. Do you know what I mean? Something so simple to use because yeah, N is is complicated. Anyway, you you were going to say so um maybe you share the the N8 you just open N8N uh io.io. By the way, did you know that N8N isn't is a German company?

I had no idea. What does it mean? because I'm I'm assuming they're taking bas the the the word that they're shortening has eight letters in between the first N and the second N. I'm just guessing like you've got you know got alley that's accessibility starts with A ends in Y and there's 11 letters in between. So I'm guessing that N is something like that. Anyway, I don't know. I'm trying to speculating. Yeah. Um, N8N stands for nodemation. Nodemation. Okay. So, they've made up a word. That's why we couldn't figure it out. But it is it's that concept of eight words in eight letters in between two ends. Um, anyway, you were going to say, yeah.

So, we're on an 8N. Okay. Oh, yes. Yeah. There are eight letters between um nodemation. Yeah. Yeah. Okay. Yeah, it is exactly what you said. So for those who don't know, N8N is similar to make.com to Zapier. This is an automation tool. You basically have a start. You say if somebody sends me a message, if if I get an email or if somebody uses the form on my website, then do a following task, anything. And what is super duper cool uh that you can do with not just N8N but any like automation tool nowadays is that you can add an AI agent meaning that previously you would have to like map like words word by word to different places.

Now you can get a fullon context. Let's say you get a message from anybody an email and you can analyze that with an AI agent have a a prompt you can have memory you can have database and then based on that you can do a bunch of like different things with it. Uh it's super easy to build like a I would say for people who don't know to start building something with Telegram. I find it super easy as a test to build an AI chatbot in Telegram. So you can have your own chat GPT inside of your Telegram chat. Super simple. Anyway, so that is NATN and they just raised $180 million uh in an evaluation of $2.5 billion.

So it's there that that's that's I'm going to send my friend cuz uh big number big number and they want to hire him. Yeah. 2.5 billion for an automation tool and now chat GPT well u open AI is entering this game. Exactly. Uh which is auto automating tasks agentic tasks and with N8N so far you can do like more I believe but chat open AAI is probably also going to uh add more to whatever they are um building right now. I'm using N8N mainly because it's self-hosted, so it's on my own server. Well, Hadner, and it's super cheap. It cost me like I believe like seven bucks, and I don't have to pay extra per operation.

So, typically with these tools, each time that you run your automation, you pay uh but when it's on your own server, you just pay your monthly rate. And if you have too many automation it you're you know you will see that your server cannot handle it. So you have to upgrade but you essentially pay uh much much much less uh in most cases. So it's a super cheap open source baby open source. I think I have a video on how to set up an N server on your own server by the way. And I'm so happy that open source is winning here. Like I mean sure make.com and Zapier they are so expensive especially Zapier is extremely expensive.

I think n cloud is 30 bucks a month. So n is also expensive if you just buy it from them. But if you host it anywhere you can host it on HNER and HNER is like so unbelievably cheap. Like you can get a server that would work just fine for four bucks. Yeah, that's good. Yeah, I think I pay six bucks or even 12 like dollars for my Digital Ocean Docker image that hosts this, but Hets know. Yeah, that's good. Yeah. Um, saying all that like N is is an interesting place because like yes, it's very very powerful. Yes, it's the uh sort of the perfect kind of storm of AI agents and then what you can do with them because you can integrate it with so many more things.

you have so much more control like you have to wait for well you don't have to wait for an integration but like Zapia for instance you wait for those integrations and then you can use the tool whereas this you can like attack you know you can send stuff to APIs not that you can't do that on Zapia especially now but n were the kind of leaders in all this so they're an interesting spot and what relevance do they have I don't know but series C at 180 million shows that they've got a lot of relevance there's probably a lot in the pipeline that they are they presented in their in their pitch decks that has people excited.

So, um yeah, I mean that's I don't know how they make money. Is it only through self-hosting? Well, not self like hosting it for you essentially. Yeah, cloud. Yeah. Yeah. And cloud. Yeah. That's where most of their money comes from. I think we me and my mate Kieran were talking about this. I think they've got that. I think they build enterprise tools as well, but um you know realistically the tool is built like you get minor updates all the time. There's no there's never going to be a massive overhaul. And that's where I was like well what is left for this tool? It's just scale. They've just got scale. I don't know. There's probably more to it.

Yeah, they probably want to build like a more agentic stuff. Maybe simplify things. Maybe automation for automations. That's what I'm saying. you know, they've got the MCP server, which I think could be improved a little bit. It's quite hard to spin up an A thing through MCP, but um especially on self-hosted as well. I think it's easy with cloud, but self-hosted, you've got to you I think you've got to create an automation for MCP to then connect with the with the other automations. Anyway, um but yeah, I think there should there could be some um there's room here to challenge OpenAI's agent kit or the agent builder to just a more friendly, you know, version of N that is just a bit more like Zappy, a bit more like make, bit more like agent builder.

Um but who knows? Yeah. And for if if anyone is interested in getting this and but you don't want to go through hosting stuff and NHN itself is too expensive, you can like one of the best deals that I've seen is through Hostinger. Like Hostinger gives it to you and they host it for you. So they do everything and it's just it's still cheaper than any anything else. So you can get uh the smallest version for five bucks a month. It's just so so cheap. You can get started and as you scale, if you have a lot of automation, it's a good thing and you probably are making money and you can scale. Um, saying that if anyone I've got referral links for Hostinger, so probably even cheaper again.

So, you get like 100 bucks or something like that or 25 bucks or something. So, you get even cheaper again. Um, let's keep moving with the OpenAI stuff because they there's more, man. There's more. Sora 2 on the API as well which is really cool and Sora 2 Pro um very expensive I've heard I think um yeah very but it's good that it's here to be honest you mentioned something about Sora to your end as well right what is Sora 2 pro by the way just a better version of Sora Sora 2 basically the pro version I guess it's more more detail more um more everything balls to the wall um GPU PU usage. So, it's 30 cents as opposed to 10 cents for a 720p image.

I don't know whether this is per uh minute or something. Uh stupidity. Where's Sora say when you see it? No, can't see it there. Oh, here we go. Audio video. Here we go. Um, yeah. It doesn't say it can't be 30 cents per v per per video. Surely per second. There you go. Per second. 30 cents per second. Oh man, that is expensive. Yeah. Use wisely. Use wisely. That is expensive. So, we can't make a movie anytime soon like every day. And that's 720p by the way. So, you you jump up to um what do you even call this? I don't know. Almost full HD. Almost full HD. 50 cents. 50 cents. So, yeah. Per second.

Per sec. Uh, yeah. 50 50 cents per second. That is insane. That is insane. That means if I want to make a meme out of my friends and it's like 10 seconds. Oh, man. That's first world problems right there, isn't it? Yeah. Yeah. True. Well, I I I do pay for memes sometimes. Anyway, uh we'll we'll talk about So, should I talk about like uh Sora to now or separately? Maybe we do it separately. Okay. Yeah, let's do it separately. So, you finish this and we give it like an update. Yeah. Well, I mean that's kind of it from I mean, yeah, Codeex general release um GBT 5 Pro as well. Maybe we can look at that.

Let's uh let's go and open router and have a look at if they even have it. GPT GPT 5 Pro they do. And yeah, so 15 $15 in, $120 out. And we thought Sonnet was expensive. Sonnet is $3 in, $15 out. So, this is an expensive model. You better be making mega bucks if you're going to use GBT 5 Pro. But yeah, I'm I'm sure this is a bring your own API key, maybe. I don't know. Um, so you're telling me it's not getting cheaper. I'm telling you the the models are that are cutting edge are getting very very expensive. Um, but you you got to remember they're absolutely getting cheaper because the amount of performance you can get that you used to have to pay, you know, again, and that's not these prices.

Yeah. So yeah, I think you've just got to you've got to change your expectations or lower your expectations and say, well, you know, you're now paying some of the stuff they they're free, but you're paying point of a cent for the same sort of performance you would have got three years ago, you know, as long as you live just live in the past, basically. Yeah. Um but there we go. Uh lots of lots of exciting stuff to to think. Oh yeah, I was going to show you the the guidelines as well. the design guidelines for the apps, right? For the apps. Yeah. So, there are brand guidelines which is which should be quite interesting to get your teeth into.

Um, yeah. Uh, where are we? So, this is about how to you how to build a UI inside of a chat experience. Yeah. Conversational experiences, intelligent, simple, responsive, accessible. I mean, you know, it's um it's quite a long article to be fair. Um but yeah, color, you know, how to use color, um typography, rules of thumb. So, spacing system as well. So, they they've they have thought about it. They're not um and I think they would because you know, it's it's a micro app within an app. Do you know what I mean? So, you've got to really be efficient with space. It's like mobile design on steroids. Yes. And this is also like it's been tested and it works in the east by the way much better than the west.

Like if you look at China, they have like this idea of super apps. They have WeChat and they have absolutely everything inside of that. Absolutely everything. Absolutely everything. Well, done. Not intended, but I'll take checks checks in the mail for that one. You can uh you can use that one at the party tonight. So So it's absolutely filled with mini apps. Uh you can you can basically have your Uber and your pizza and pay your friends like a share of a meal or whatever. You've just unlocked what chatbt are going for to be honest. So the there are like I've seen some videos I've watched some videos on this uh and I find it fascinating because this is about culture and also design like in eastern cultures apparently and I'm kind of like in the middle you know I'm from Middle East so it's um you're literally in the middle.

Is that another pun? Is there a pun there with that? What is happening today? So yes, so uh apparently like in Japan and also like in in China people care they want to see all the information in one view. So their websites are more crammed with every piece of information and on the west it's much more like separating things from one another. That's why this idea of like super app has been tried many times in the west and it has never worked. Hasn't worked yet even like Facebook tried that. Facebook tried to do everything in one place and not really working. Uh ex you know Twitter like Elon Musk bought Twitter to specifically try this.

He's still kind of trying to like merge everything uh in one place but it hasn't. Telegram is kind of like the best thing that has been it's not the super app but it has like a lot a ton of features. It's so featurerich and it has like a lot of users I believe like over a billion and it's also popular in the west as well. So I don't know if because of that culture reason if uh open AI the idea of like a super app could um could be successful because that there is also culture uh incompatibility with it but yeah we'll see I don't idea I don't I don't Yeah but if it's good if I can if it can make my life easier in any way like I find a a lot of times that I want to like get help in curating.

I'm not a music guy. I don't listen to a lot of music, but if I want to do I want to get some help at curating like a playlist for me, you know. I do actually get uh I I ask my friends and like my siblings for like to send me songs because I feel like sometimes I feel like outdated. I'm like, "Okay, well, can you send me some good stuff?" I don't listen a to a lot of music. I'm more like podcast and anyway. So, we'll see. We'll see. We'll see. We'll see. Cool. Let's move on to plan mode in Cassa. Uh, wherever it is. I think I've just got a video. Yeah, I saw this.

I saw this by accident. It just I I opened my cursor and there was like the plan mode and I was like, "Oh, cool." Yeah. Um, so stop raw dog in your AI development. Plan mode is finally available in cursor and I am here for it. So yeah, it is exactly what it says on the tim. I think they introduce a few little nicities here, but if you're not familiar with clawed code, which already has a plan mode, you shift tab, which is the same thing here. For me, this is the recommended way to build features. I think that the idea of a plan mode is when you when you send a agent off to do some coding, it's kind of figuring it out in real time.

It kind of broadly looks at everything, but it's it's almost like shoveling snow while it's still snowing. It's making changes. Whereas this will just look at everything and plan it out. And if you can combine this with some MCP servers, if you can combine it with some uh level of like thinking mode and whatever, then you'll have a plan that it outputs to a MD file that one you can reference later on. I often do that. I'll make an MD file so that I can push it up and then I'll do that in different branch later on or whatever. Um but yeah, it will give you a plan before then you start executing it. And what often what often happens it breaks into phases for some reason tells you it's going to be like this is week one and this is going to take a week to do and this is going to be three weeks like it still uses antiquated um what do you call yeah so but I literally built a sixmonth project in like what do you mean you aren't you going to oneshot it for me like in 30 seconds what do you mean six month okay so If I say it in a different way because I noticed this but I haven't used it but so the the way I see it essentially today when you use cursor you have the option to chat or like use the agent.

So the agent does something out of like just right away and it might break things. You don't want that. And the chat I noticed that I go back saying okay the things that you said should be done now do them. So I'm going like in this mode that should have a in between mode and this is exactly that. That is kind of like chatting with the AI but what you get is not just a chat. What you get is a plan that the AI can execute on in a step-by-step way. And this gets me excited because last week I started with um like a GitHub repo from somebody from the Webflow community to to essentially learn what this guy is doing like some crazy with animations and I want to you know get in and learn it.

So I I'm trying to I I was trying to learn learn that using cursor and it was not doing what I wanted because I wanted to give me a plan to look at the entire doc the readme file everything and give me a plan how I can use this structure and start like building piece by piece but it the chat mode was not good because with the chat mode it gives you a plan but it's it's in a chat at and you can't go you can't say okay number one is done let's go to the next one but you could you could say output this plan to a MD file but now we can now we can that's what I'm saying well you couldn't do that before well I didn't know how to do it well you didn't think you just didn't think to ask this plan think for me but but you know you again you're the master of your own destiny so just correct me if I'm wrong but it sounds like you want to implement this like you want to write it.

You don't want the agent to write this code. Yes. In this case, you plan. Yeah. I mean, in this case, I want Yeah. In this case, uh this repo is about me learning. It's not a framework, but close to like a yeah like a mini framework how to do like cursor and g like not cursor, how to do like gap and uh 3JS and this kind of things inside of web flow. Yeah. with you know with this framework and I don't know a lot of these things so I have to start from the very basis but I want to do it following like this guy's experience I mean there's nothing stopping you going into chat GBT and again master your own destiny tell it that you want to implement it and that you it needs to tell you what code to add and why you're adding it and what it does right so you need to tell it that there's nothing stopping you doing this inside of chat GBT but the benefit here being inside a cursor is that has context to your codebase.

So this is how to implement it in your files that you already have. So this is this is great. Um you know I always struggle to understand why this is any different to for me this is just a a system prompt a a ch you change the system prompt. I don't know why this would be any different to being in edit mode but you say do not make any changes just make a plan. I I don't know what the difference would be. It would be interesting to find that out. But it's nice just to know because I've, you know, I guess in some instances I have had it just start working on stuff. No, no, no, don't do anything like I want you to just whatever.

Yeah, that all being said, you got to tell it that you want to implement it, but it will just Yeah, it will just look at all of your files and and make a comprehensive plan for you to do later on. And this one automatically makes an MD file for you. So I often have to ask claw code to make it an MD file for me to output it to it. just creates a plan and that's it. Asks me if it wants to start implementing it. The other one is um it'll I don't know whether you've noticed this, but you'll it'll actually autorecommend itself. So, if you're writing something that sounds like you're asking for a plan, it will say, "Do you want to use plan mode?"

So, that's quite good. That's quite nice. Um and then it always comes back with questions as well. So, that I like that feature. oftentimes claw code almost just never come back with questions trying to like clarify things. Maybe I should add that into my root memory file to say anytime I'm in plan mode, ask me clarifying questions because yeah, it's um that's a nice feature that I saw on this as well. So yeah, pretty good. Yeah, and I would recommend it to everyone building out big big features like just take a moment to plan it. and take a step back. Just be methodical. Be I know it's we're so eager to like just build build and like work fast.

Like take your foot off the gas. Just be a controlled organized freak when it comes to AI coding. So yeah, pretty good. Cool. Uh I will I will use that. Actually, I'm going to restart this project and use plan mode to to basically learn. Uh, shout out to Fedrico for Fed I believe like I is Fed Fedrico. Yeah. Um, Fedrico Valla for his amazing work by the way. And I don't know if he wants me to share this or not because I got it as a Twitter DM. So I'm I'm hesitant sharing it. I think he's he has plans on like announcing it uh publicly. But if you search it on GitHub, you'll find it.

So if somebody really wants to work with his um framework that he's working on, you can find it. So yeah, my mate is texting me about uh N. So he's he's making making tracks within there to be um to whatever because yeah, we spoke about everything we spoke about just before simplifying the experience and whatever. N are perfectly primed to be a really successful company and um I think it would be a smart move for him to go and work there. But yeah. Yeah. Um, so let's move on to Anthrop. There's anything else about cursor you want to talk about? No. No. Um, Anthropic, uh, think they're the big cheese when it comes to AI alignment.

So much so that they've just open sourced their internal alignment tool for anyone to modify. Now, this is quite nice. I think it's a bit uh, how do you call it? egotistical, but yeah, but that's just I think I'm just combining my vibes from um anthropic more than actually what this means. This is actually a really good tool. So, we don't have to talk much on this, but Petri, an open- source auditing tool to accelerate AI safety research. So, we've discussed alignment. We know what alignment is. um acting in within the best intentions of humanity and then actually does what it tells you to do. But what I found quite interesting is the open source nature.

Not only is people are people are people allowed to use it for free, you're able to modify it. So the UK AI security institute have modified it with their own preferences. What what do they consider alignment and what it's an they they've changed it into an auditing tool. If we click in through into here, see what it says. Um, to make sure that well AI governance is done correctly in their institute. So I quite like that. I quite like that you can I'm presuming you can run any coding tool through this to be honest. You can say look this is what we it's a yeah governance and and uh what's the word I'm looking for?

compliance mechanism to say this is how we deem an AI to be workable and whatever within our organization. Petri have already kind of you know taken the lead on that. So quite nice little I mean it's always good to see open source stuff but yeah it's kind of um that's kind of it. It begs the question who will adopt it and um again anthropic have their own ideas of what alignment is and whether everyone deems that because I think a lot of people will just use this at face value and and won't modify it whatsoever. I mean there's no harm in that but you're taking anthropics trust which we are already doing anyway right open auditing their have got their own alignment um goalposts and all this that and the other so yeah pretty pretty cool anything to add to that not much I'm not sure yeah um yeah it it's kind of like nice to see but I'm not sure honestly yeah um we also have uh a really interesting move.

We talk about the you know the app store and stuff like that. Really interesting move by Anthropic again where they have uh in they've potentially just had their app store moment when it comes to core code plug-in marketplace is now available for all of your slashcomand agent MCP and hook needs. Now, if none of that makes sense to you, um, basically it's an encapsulation. Like, claw code has slash commands which enable you to use reusable, what do you call it? Reusable prompts, reusable system prompts. Like I have one called question, which is similar to what we were talking about when it comes to it just says answer the question, don't make any changes. So, I have a slash prompt that just always does that as an example.

But um yeah, you can write you can create MCP servers with the with their new agent SDK as well. This could be ones that you've written or claw code specific ones. And then hooks are they're actions that respond to certain stages in the the agentic workflow. So something I want to look more into. I want to get I want to nerd out on that one weekend, but I haven't done so yet. And so this is an encapsulation of all of those things that you can then provide to the community. So I actually thought I could demonstrate or even just not even demonstrate sorry but just look at it practically um with with regards to if I stop my screen share inside of the context of core code.

So you can actually we can all see it. Um where is warp? Here we go. So basically inside of claude code. So if I type cool, let's clear this. If I type claude, it's going to load up do its thing. If you type slashplugin, you you've got to add it looks like you've got to add the marketplace. So I would and these are just GitHub repos. So if I my marketplace is like a git github repo. Um you add the marketplace and then within that marketplace. So it's adding it here. Okay. So it's already installed. Very very good. And then if I escape that then within here in fact I'll just do it again. You can type plugin and then um browse browse and install plugins.

So I can then I've only got one installed. So I've got the call code plugin thing here and then I can look. So I'm interested in the feature dev one. So it's one it's this is the one that they've used internally that I'm guessing that whenever they add a new feature they've got a set of MCP servers, set of prompts, set of this and set of that that mean that their features always go out with this level of due diligence. I don't know this for 100% certainty but then yeah it makes sense also security guidance. So quite nice that you can this could this be something that people sell? I don't know like is this something that you could not not sell sorry but like you know what I mean make a make some sort of impact like the web flow marketplace as an example.

Is this someone like oh we just build we just build um plugins for claw code you know. Yeah but it's super niche. Yeah right. Yeah. Yeah. Very very niche. Very niche. I'm quite excited for it. I'm I'm still trying to wrap my head around it, but that's that's my understanding of it all. Um, yeah, be interesting to see how how that goes on. Um, let's get on to some framer stuff, I think. Do should I open these and I'll share them or you going to share them? Well, uh, I can share. Uh, the thing is we wanted to also talk about Sora a little bit. Go for it. Yeah. So, Oh. Oh, yeah. I was going to be going to bring that up during the the other segment.

So, we'll have to edit this. So, let me um Yeah, that's it. Yeah. So, so we talked about Sora also last week. You've probably seen some Sora videos. These are like AI generated videos. The company behind it, OpenAI, they are losing money on it and they have a huge financial problem. Um, it seems like making funny AI videos is not generating a lot of money for them and it costs them a lot of money to make those videos and people are not really paying. Uh, and people are also not using their app because they have a social media type of app with it that people don't use. They just generate the video and take it out and share it on social media somewhere else.

So they some altman the CEO also had this blog post on his uh personal website saying that they are going to make some changes. Uh one of them is about uh copyright. So they had this thing that was very backward like copy uh right you had to opt out meaning that if you didn't want your content to be used you would have to contact them and basically tell them do not use our content for example Disney would have to do that and now they have some issues with uh Spongebob and I don't know like many characters that people are using and yes they are Nintendo yes people are the audacity of this, the audacity to be like, yeah, and he knows.

He knows like, of course, he it's not dumb. He knows exactly what he's doing. He's if he feels like he can get away with it or just with a fine with a slap on the wrist, he does it, right? It's not probably won't go after him. Do you know what I mean? Yeah. And like it's gone gotten really bad because people generated I don't know Spongebob with drugs things like this and of course um Disney was not happy and many examples like this is it such a debate is it such a debate yeah and they're not making money on it because exactly that's the that's why that that prompted me to say it's such a debate because these people aren't making money out well actually who are you saying isn't making money what you talking about the Teletubbies aren't making money out of this nobody Nobody.

So nobody is No, nobody is making money. Technically they are but um uh not really. They are no they are not making money. They are getting some money but uh like the money that it cost to run those servers and to to to basically get the AI to that level is costing them far more than what users are paying for uh whatever they are paying. So yeah, at the end they are losing money. And there is this good example that came to my mind like comparing it to this meme generator. I shared this on on the stream before. Uh I think you've used it as well. Let me zoom in on this. You see this?

Sorry, I'm just sending scumbags to a a Telegram group that I've been added to that they're talking about trading and whatever. Sorry. You have to opt out. Yeah, I don't let anybody leave. Anyway, sorry, I'm back in the room to me and meme generate. There you go. There you go. So, so this tool this tool doesn't make money. Well, maybe a little bit with ads, but not really. So, meme generator, we use it for memes like on our Discord. We we use it all the time. Okay. Uh, and what this is, this is essentially we have some like internal jokes with friends or like we make fun of a friend and this is how we do it with a meme.

We put each other's heads on characters or write each other's names depending on the situation. And to me, honestly, like Sora too today is just that but animated and like fun. So yes, there is this idea that you can create like advertisement and like movies and like crazy with it and theoretically get paid for it. Yes, you can do that, but it's the quality is still not there to be that good in most cases. Yeah, in some cases you could, but in most cases it's not there to do everything that you want reliably. Um, so what people ended up using Sora for is just essentially creating memes that is a glorified uh version of meme generator and it's causing that was always it was like you know Sam let his what do they call it when you can you can let your face be used as it like a Yes.

Yes. I I know the the toilet one, but yes, the meme was a part of it, but we were promised uh AGI like we were promised, you know, cancer treatment. And what we have now is quite literally a bubble. Like I don't know how else to call this if not a bubble. And as you mentioned it earlier in the show, maybe it won't burst. maybe or even even if it does maybe the burst is not so huge to take down you know the society or like any industry with it but it's probably going to burst because the way it is now it's kind of like very funny Nvidia is uh giving openai money and OpenAI using that money to buy chips from Nvidia okay and OpenAI is buying chips from AMD and AMD is giving OpenAI 10% of their stock.

And now we have Oracle somehow that is going to use AMD chips to build something for OpenAI. OpenAI promising to give them money. You see it's all a selfperpetuating situation where a few companies are you know taking the shareholders money essentially because when they talk about it the the value goes up they are taking that money and they are putting it back in each other and it's just this loop and the issue is who is making money like none of the okay Nvidia is making money. Nvidia is absolutely making money but who else is like Nvidia is not uh OpenAI sorry is not actually making a lot of money. Oracle we don't know how the deal will be.

Uh but a lot of the focus is on open AI releasing something promising us and releasing it and delivering it and the idea that if they do they will make money on it but they haven't yet and I'm not saying they will never but they haven't yet the they are making money but their costs are significantly more than the you know the amount of money they are making and they are running really hard trying to get to the point that they can keep the users in because open source is growing as we saw with N8N and we see with AI uh you know um models there are not a lot of incentives to stay indefinitely with open AI so all I'm saying is if AGI comes along if like these agentic things get like truly Amazing.

Open AI might not be in the best position to to to get money from users. And all we have then is, you know, a bunch of deals, a bunch of chips and warehouses that cost a ton. And then Nvidia in this case is yes, they are making a lot of money and it seems like they will be they will continue making a lot of money. But if this is a bubble and if this bursts, Nvidia is going down with it as well because then people won't buy like companies won't buy their um extremely expensive chips anymore. They won't be able to buy because nobody's paying for their services. And right now it's it's just like paying each other and the share goes up and more money pull pull uh at the end of the day though by the way just more importantly 69 viewers.

Hello. Um oh no we're up to 70. Oh no somebody leaves should leave. Question is why do we care? Why do we care? Yeah. uh if we are in a bubble and if it bursts bursts uh we'll well a lot will go down with it if in the best case so uh I I believe it was Jeff Bezos talked about it we we have like two types of bubbles that we could be in one is like the financial crisis in 2008 if we would have something like this the whole society will not collapse but will have challenges will have like difficulties with jobs, with money, with everything, right? Day-to-day life will become difficult at least for a few years.

But another type of bubble we could have is like the dotcom bubble. So the.com bubble is the early internet bubble where there were a lot of um we had a lot of startups uh trying to do something with the internet and they had not they had not figured out at that time how to make money with it. So they were like internet startups and investors were pouring money in it but nobody was making money kind of similar to what we have today. We knew we knew for the fact even at that time even people who were like pessimistic about it they knew that the internet and that you know the idea of it is amazing and we will eventually make money out of it.

It's connecting a lot of people and services and companies but we didn't know how. It's the time before like Google. I I don't know if it was before Google but uh I'm sure it was before Google making like money anyway. So it we had this com burst a lot of companies um went down essentially they went to zero uh well they went to minus like they they lost all the money that investors uh put in them but a few came out of the ashes and rose to the top such as Amazon. Amazon one of was one of the survivors who become a giant. Now you could say well even if it's a bubble and it bursts and it's not the end of the world but actually at that time a lot of people a lot a lot a lot of people lost their jobs a lot of people had difficulty so it's not going to be easy that's one reason another reason is again I'm I'm I'm repeating myself here but we were promised AGI I want to see AGI I don't want to see a bubble I don't want to see I honestly I don't like seeing Sora, yes, it's fun and I would use it for meme generation, but I would like to to see like true AI agentic that can do things and not this increment that we see here, like this increment improvement that we've been seeing in the past three years.

That has not been really what we maybe maybe we're just not looking out for. I'm sure, you know, I'm sure we can find some very optimistic sounding articles like we're we're we're just a bunch of kids. I say we're a bunch of kids. We're not really, but like, you know, I bet there's some pretty optimistic looking articles about um AI. You know, we talked about the um mind password thing. We talked about like there's going to be stuff going on, but how do you make money with these? like we're talking about an industry that costs a lot of money and we we still don't know how to make money with it because uh I believe it's like 3% of the US uh electricity is going to AI that is insane like the amount of uh the the new deal that I believe I don't know which company was the the new deal that they made to generate energy is like enough energy for 5 million houses.

It's the investment. We say the energy thing. I I actually watched a Dave was it Dave Shapiro? He echoed my sentiment. I think the whole economic e economic I think no not economic ecological impact of AI isn't exactly like yes it can power all these houses that's fine but how many like where is that got? It's probably taken that energy from elsewhere right? It's not people aren't driving as much people aren't spending so much time on their whatever it is. Anyway, that's just yeah and and there are unprovable ways to make energy and it's kind of like a different topic but I don't know are you familiar with the card I believe it's called Kart scale.

Yeah. Yeah. And we we're not even close to level one societ uh civilization and we want to go there. I'm actually looking forward to have reason like proper reasons to use more energy and build more ways to generate more energy right because it means we as a society as as civilization we are advancing right I like that I like that idea but I don't want that to be used for Sora to to generate toilet sin in the toilet videos let people have fun. Why are you such a pie pooper? Let me They're fun. I I will I will contribute to the fun and will make some videos like this out of my friends once I get access to buy.

Yeah, got to buy your um How much are they going for on eBay? Did I ask you that? I don't know. I I think I asked Chris if he can find me a deal. Only and people are selling their invites on on eBay, which is um Yeah. Anyway, let's let's move on. framer pricing. Let's move on to framer pricing. So, uh if I show So, framer has a new pricing announced and you probably won't like it especially if you were looking forward to use framer or you are a framer user. So to use framer to be like exact to use framer as a designer it's free and that is amazing about them. Another great thing about their pricing is it has been simplified.

It is super simple to understand but it's quite expensive. So they have this basic plan that cost €10 per month that is not useful. you you cannot use this for pretty much any website except like a simple portfolio. Um it it gives you one CMS collection which is better than nothing but it's not really useful. Like you can't you can't do a blog with a tag. That's insulting. Yeah. It's like it's for a blog. That's all there is. No, you can't. Like even for a blog, you have authors, you have tags, and if you want to build something proper, you need for a blog, you need like no blog. Yeah. Like it's I don't know.

It's really Yeah. Yeah. Okay. Okay. So, if you want to to build something that simple, you shouldn't even pay 10 bucks. You could just build it in Web Flow, export it to Netlefi, whatever, and you just don't pay anything. Anyway, the the basic plan doesn't have staging, which is uh zoom in a bit. Staging and instant roll back. Yeah, there you go. Nice. Well, I don't understand how that can be used. You don't have backups. Anyway, so that's the basic. Then we have the pro, which is uh the the pro is even worse. You have €30 per month. So So that is quite an upgrade. And if you pay annually, if you pay monthly, it's 45.

So that's that's quite expensive if you want to just, you know, get in and try your website for a few months. That's very expensive. And here you get only 10 CMS collections. And why am I saying only? Because I went back I I checked my client websites. the the websites that we have launched this year for our clients, none of them have the less than 10 CMS collections. So we cannot practically we cannot use this 10 CMS collections is not enough because you have testimonials, you have like your blog, you have your authors, you have people to add in it adds up quickly and you have like a fancy navigation for example like for one of our clients we have a fancy navigation and we want to make it as easy as possible for the client to manage those uh items in the navigation.

We use the CMS. So the CMS is not just for a blog. It's no, it's used for any repeatable like content type that you have that you want to kind of like use globally like your FAQs, um your glossery, so so so many things. Anyway, so this is quite expensive. Now, that's not the end of it because uh you get only 10 CMS collections for 45 bucks a month. If you pay monthly, you get 2,500 items. It's uh okayish. I'm I'm not Yeah, that that's okayish. But if you add a editor, somebody else who can with another account join the project to edit the project, it cost another 50 per editor. Are you kidding me?

God. Are you kidding me? Where are you coming up with these numbers? Like why? Like what is the justification behind this? Like is it like the platform costs? Like what what what is the justifica? Like you can't just make up Well, I guess you can, but this is just made up prices. Um anyway, translations cost 25 per local. And by the way, these are like annually uh monthly. If we do annually, it's a little bit better. 40 bucks per editor. It's still really bad. Most people say mo most people are thinking you think of your website in multiple years. You don't think of your website in multiple months really. Unless you're playing around, but like yeah, it's still it's expensive.

It's expensive. And 20 bucks per local, per like translation. So if you add another language, you pay €20 per month per language. It adds up quickly. And now that's not it. The problem is that the pro is so small that it really fast like actually fast for any client project. It pushes you to the next one which is scale. It's a hundred euro per month per website and it's annually only. So you you're paying €1,200. Yes. 1,200 for your website. And just with that, you now have like custom local regions. You have like AB testing. You didn't have that before. and you have now you have 20 collections and even that is to me insulting that you pay €100 for 20 collections with web flow you pay I believe you pay like €30 for like 40 collections it's just much better like with web flow so this is what I said like contrary to popular belief web flow is actually much cheaper than framer webflow.com I will make a video on this like going in more details.

But yeah, what do they know that we don't know? Because they can't just be pulling a putting their finger out into the wind and thinking $50 per 23. See $23 not euros. Yeah. Yeah. And you get here you get 20 collections. This is this with framer this cost 100. It's like four times over than four times more expensive. Actually, like if you do dollar to euro, it's five times more expensive. So they say, you know, unlimited Figma designer or frame of designer, sorry. Um, design pages. Yeah, unlimited pages, unlimited uh, you know, whatever. But they're making it up once you then, you know, um, what's the opposite of churn? when you when you promote yourself to a paid plan because their paid plans are crazy expensive, but then it's like, well, you get the designer and that's that will be their way to justify you get the designer for €100 per month.

Do you know what I mean? Yeah. Yeah. Yeah. And what I like with Web Flow now, they have this they previously had this thing that you would go from CMS to business and then you would you would be pushed towards enterprise which is expensive in with every company. Um but now web flow allows you to basically choose the amount that you actually need and independently like if you need more CMS items you buy more CMS items if you need more bandwidth you buy you need to combine the fact that web flow charge per site per with organization right there's two types of plans so what is what is the total but it doesn't matter for your client because your client doesn't pay for workspace so workspace plan is for agencies this is essentially me if I want to have a big team I pay for I do I do pay for for the team functional most agencies sorry so most clients should probably upgrade to a plan because that you know most of they're more than happy I always found when I worked with big clients they had agency plans or whatever because they would have little they would have landing pages and they would have other websites you know they wouldn't just have one website.

Yes, in that case um they will so I have some clients who do pay for workspace plans but most of like mo like 90% of my clients on the free version so it's an extra 20 bucks for that and and so do frame would that say let's say for that $40 per is that the workspace as well with multiple sites that's that's the good thing that I mentioned that uh earlier on that you don't have that so that is free But if you add editors to a website, it cost again like €30 per editor. So it's kind of it's different, but it could get more expensive even than the web flow workspace because you are paying per editor, which make sure you do that in your video then.

Yeah, because like you know you've got to take into account that you got to look at your current audience uh sorry your your customers and say well how many of them have multiple designers and developers because you know um if you only got one if most of them only have one then you don't have to worry about that $50 extra or what or 50 euro extra but um there is a total that we need to take into account but it does feel like that framer are just a bit more expensive but then you do get the designer that is a selling point to be honest. Well, here's the thing. If you say Web Flow has like this agency plan, this is for your workspace.

So, you would have you even if you pay this, you pay this for your entire you know for per for that person for the that that's like access to the team and then you would have unlimited projects etc. And then when you what do you do with those projects? With those project you're going to go live. So web flow is going to be much like much cheaper when it comes to actually utilizing the projects that you have on your dashboard and setting them live in because with framer you would have to add editors. So framer kind of like to me it gets not exponentially but it gets much more expensive per site that you add because the individual sites are so much more expensive with web flow.

Yes, I agree. And this workspace plan like I I think Webflow should get rid of this. I don't know their numbers, but I don't I can't imagine they are making a big chunk of their money from this because this is only bigger teams who pay for this and the if you just look at just imagine the numbers I assume we have by far more live websites than teams. Yeah. Yeah. Well, yeah. I mean, I look forward to the video and seeing all these looking at all these different use cases because there are several use cases. Do you know what I mean? So, it be interesting to see who comes out on top. Do three different use cases and see who comes out on top of each of them.

I can imagine that. That's a good idea. Yeah. Yeah. So, um I'll do a more in-depth video. Uh but yeah, this is this is still quite expensive. I I need to look into the community what the framer community says. I because I actually don't know. I haven't checked Twitter today all that much. So cool. Okay. So the next one um I have honestly like I have three things that we can bundle together. Great. Yeah. Yeah. So a few other things uh in the web design and web development that I want to share with you that happened this week and I find them to be very interesting. One of them is this motion. Let me actually repeat this and like zoom properly and then repeat it.

Okay. So the first one is this motion design made by Gal Shear. I think I pronounced his name wrong. Uh, and it went super viral. Let me actually reload the page. It says 1.5 million. It went super viral. Like multiple people also shared it. And it's a banger. Like look at this. Is there music to it? Yeah. The sound effects, the sound design is also really nice. He did this branding for Framer. I don't know if he did it for free for just just did it or they asked him to. But yeah, this is so sweet and this is done with After Effects. So the I I saw this post. I saw it on Twitter like 10 different people shared it on Instagram.

Uh I saw it. If you do like cool work, people will share it. Like including us, we are sharing it because it's it's so exciting. It's so amazing like the work that he done. So that's something cool. Uh speaking of cool work, another one is how was the webflow.com like the the new web flow website, how was it built? So Corey who is on the web flow team, he's a designer developer, he led this project and he shared in this video um with no code supplied that that's like his uh tool that he's building. He shared some detail about how the website was built. What is interesting about it is that uh it was so the project was like 15 different web flow projects mixing them in one.

So they brought in a lot of like different separated projects in one uh project and like I believe like mixing 45 different um CMS collections and like just bringing things down and making more lean in one project using the new component system uh and also using the real time collaboration which we still don't have access to yet but they've used it to uh have six developers on this project at the same time building this giant this is like an insanely huge website all with components and they used mast mast is the framework. Yeah. Yeah. Yeah. Uh so they used mast and he shows how the component system is used and interestingly enough it's not a lot of different components because with component variations and with component um properties you can create multiple layouts all with one component.

And to add to that we have slots which allows you to put components inside each other. you have cards and you you just reuse them in different places in different like versions and these render only the piece that is needed. So they don't like you don't get the entire junk in the front end when you load it. So it's super performance. It's not just showing and hiding things. It's literally rendering only what's necessary. Exactly. It's not hiding it. It's just not publishing what you don't use. So, this video is a cool shout out um to watch to see how a huge project like this is because it's seven minutes. I wish it was longer. Yeah.

You sort of want it to be like a series. Do you know what I mean? I want like a twohour version of this. Like honestly, I want two hours of this goodies. Yeah. because it's quite it's you know maybe not so much for the design but for me it's very interesting thinking the strategic side of you know when I first started doing web web development we would literally print out the website right and we would circle things and we'd create like a mind map not even a mind map like a physical map on like well um what's the word semantically this yeah component is the same as this component they just look different so how do we do we combine them do we create variations like you create a whole map of how everything is and I'm sure they had to do that to get the most out of as little as possible, you know.

Yeah. And look how lean and uh Yeah. Look at that. The the variable panel is just so lean. It's so little. It's not like crazy efficient. Yeah. It's not uh crazy. That's why I love mast by the way. I I don't first always wanted to do a video on mast. I always wanted to like look into it and whatever but I just never got round to doing it. Yeah. And because like the the entire community started like using uh client first we have to use it but I personally like mast and mast is more developer friendly and leaner and client first is more I don't have you tried lumos. I've watched videos on it but I haven't tried it.

To me, Lumos is very exciting, but it feels a bit and I don't want to use the word experimental, but it a lot of it looks very complicated, honestly. And I am afraid of like trying to use it and not I I'm sure I will understand it. I don't think it's that complicated, but I have a I don't feel like great sharing something that complex with everyone else and like handing that over to the client. Maybe I should rethink this. Maybe I should give it a try. And I wonder what his and all of these people what what Cohen's um my is it Mory Cohen or Cory Men? Cory Cory. I wonder what their their oneliner is for these like this is the this is the design or these are the this is the style framework for you know we can whatever we can look at this uh so mast is a lightweight framework of essential classes components and strategies to help you build efficiently for any brand and scale in web flow they are efficient and they are web flow uh they are developer uh friendly and they are based on The idea of mast is based on bootstrap.

Interesting. Are you familiar with bootstrap? Yeah. Which is very interesting because bootstrap is is kind of dead and it was bloated and horrible. Yeah. Yeah. But you know laid the foundations of of a lot of what we do today. The componentiz you know components really started with bootstrap. Yeah. So try to do less um try to do more with less CSS. Anyway, I again I like it because it's more developer friendly um and more leaner. Uh if it's leaner at the start, it gives me more freedom to to add whatever like system or class that I want on top. This sounds like a good idea for a video for you as well though is this idea of like taking mast taking Lumos, taking client first and saying this is what it is and this is why it's good.

Um who who was that mustel guy? He small, medium, large or something. He had one that's um that looks like he's working quite heavily on that. Have you heard that one? No, I'm not sure. Like I know there are like other frameworks as well. Yeah, but just like again establishing those oneliners based on your experience like this is great for do it's a lot of research to be fair but would be very helpful because people probably are looking at all these frameworks and be like well which one is good and why would I use this one? How do I know it's going to be right for me? Yeah, I think to me to me masters it seems like to be the middle ground that is lightweight and simple enough that they can hand to another developer and understand it.

uh client first had this idea to share it with your clients, but I never I understood why why why would I want sorry fu team but why would I want to share like the designer the classes with my client they that's that's amateur designers which we all were at one point thinking this is a good idea thinking oh yeah they want but then you get more advanced in your career and you learn that No, like we do not want they they they shouldn't want to. They should focus on what they're good at, not and they're hiring me to be anyway, whatever. And and web is nowadays amazing at being good for um like low and non-technical user.

So as a developer, I can build this beautiful like component system. These green things that you see on the screen, these are components. Yeah. And you can now put these inside of each other and any non-technical people can do this because you have a component that is a button and you put it wherever you want. Well, not exactly wherever we want you want, wherever I as a developer allow you to. So there is less chance of you breaking the design. And if even if you do, you can it's like Lego blocks that you can pull out and like put back together any how you want. So yeah, client first. H I'm not a big fan honestly, but we use it.

We use it because the the the entire like ecosystem. Anyway, so the last thing that I want to touch on is about this developers love headless until they'll uh they are stuck maintaining it. Uh and I feel like you will have some feelings about this. So the whole idea is should you use a CMS like web flow which has API and you know you can also build something around it and on top of it or should you go just completely headless and build your own CMS with a headless solution that you have to maintain and update and scale. So yeah, well case by case basis, but yeah. Yeah. Yeah, that that that's a good way to put it in.

But the way they showed it, by the way, is like with headless, you have editor and then the headless and you do the web hook, the localization, AB testing things, and then you do Mhm. you know, the builds and then until you get to the live site and you have to handle the caching, etc. With Web Flow, you have just the CMS, you edit visually, and then you publish, and that's it. U I mean, and they have this comparison. My thoughts are building up. My thoughts are building up. Yeah. So, I mean, what what in a nutshell, what's going on there? I can't read that at all. Uh, okay. Well, architecture, content, modeling, uh, hosting.

Well, yeah, of course, hosting, but versioning. So, what are they comparing this to, by the way? This is looks like a comparison chart. Like, what are they? They're comparing it to headless to any headless, just general headless, any other headless CMS like Contentful, Prisma. Yeah, exactly. Exactly. and versioning like you do have to have a requires git based but web flow has built-in branching and staging and roll back sorry doesn't exactly because web flow has backups but it doesn't I don't maybe this is enterprise so can you can can you go to a website and just buy the web flow headless CMS is that what they're saying here well you could You could do that.

You could buy the Web Flow. You could use the Web Flow CMS and then, you know, use the API to to use the content anywhere you want. So, this this is kind of forming the the backbone to my opinion here. It's like unless you can buy this separately and they offer it, you know, all of this stuff for, you know, $5, um, then cool. Starts to look a bit more appealing. But if if you can only get it when you have a web flow website, then really all this says to me is the only people who should use this are people that want to share data between their web flow website and yes, an external app.

Yes. And I think that is also what they are realisting look realistically looking at. So it's not really comparable to all of these other tools, right? Not necessarily. Not in this really good. This is just a really good feature to know that you can actually I I am surprised that you know you can you can read the web flow I thought you could read the all of this stuff through the web flow API anyway. So I wonder what you can but uh I believe um I don't know if you have like some limits on it. Well, you do have some limits. If you if you use a uh an um a free account and a free site plan, then you have a limit on the amount of CMS items that you can create and the amount of collections that you can create.

Sure. So, how is this different then from, you know, whether it's free or paid? How is the headless CMS different from just me accessing the API? I don't know there I don't know I haven't read you should have no I I just saw this and I was like I haven't read the entire thing I skip through it and I don't understand everything that's the part also the a part of the issue I haven't worked with any headless CMS yeah so I wonder if there's a UI that you can bolt onto let's um I don't think So I don't think so. I think this is meant to be used just like the the way you use web flow like through the web flow designer.

Um just um headless CMS definition head of until they stop maintaining it. uh if you're valid makes sense API you get API structured content model freedom do what you want but websites aren't just developer see visual tools offer a faster path but teams don't often feel forced between control and velocity web flow takes a hybrid approach and visual first CMS that might yeah so it's it's the interface that you already know and love that you naturally use and they're just revealing that through so I just I just I'm struggling to understand what the differences because as a developer it doesn't change anything like no one's going to use this headless CMS on its own. It's all it's it's most likely going to already be part of a website.

So I just exactly I just I'm sorry Web Flow, what are you talking about? So why a hybrid CMS works better? They have it here. But this always existed is my point. This has existed since the first Web Flow API. There's nothing new here. Yes. What they are trying to do is like convince you why you shouldn't use a headless CMS because you have to maintain it. Okay. It's kind of like that that's the point that I got across. Sure. But why would you pay all that money? Like where's the pricing comparison between using Contentful and Web Flow? I mean, they should have. Is there a pricing comparison anyway? No, because like what? Okay, let's take what is is quite expensive to be honest.

So, um maybe it is cheaper to literally just buy a web flow um buy a web flow thing. $300 a month, free forever, 10 users, two rolls, um 50 gig. So, free is pretty generous, but then it goes way up. This is Contentful, by the way. Um yeah, but this is this is pretty good. Like uh this is live collaboration doesn't exist in the free version. Mhm. Okay. Commenting doesn't ex you can't schedule. Okay. So you're pushed towards the light version which is quite expensive. Yeah. Contentful is definitely for like big teams and with big yeah CMS requirements. I'm literally asking Comet AI why would I use this over Contentful on the flow. Um, you also have Sitecore.

Do you know Sitecore? Yeah, Sitecore is a a WordPress basically, but they offer a headless solution. So, the pricing here, I don't even see a pricing. Uh, come on, comet, what are you doing? Okay, just bailed on me. Why would I use Webflow headless CMS over contentful? They don't have a pricing. I I have no idea what Yeah, it's like it's like enterprise like Yeah, when they don't have this, it's like, oh, okay. It's just 100k like starting at 100k. So exactly. And businesses think they're pay they're pay they're paying for the SLA. They're not paying for the technology. They're paying for the the 100% uptime and this and that and the other. Why aren't you working?

Anyway, uh this is kind of maybe something that you would like to go through sometime. Well, I'm just I'm still I'm waiting on my response from uh Oh my god, why did I do that? What? Got to do it again. Why would I use Webflow headless CMS over contentful? There we go. If you look at this diagram as uh also we have the web flow designer and then you can do the you can use the API to have your web app and you have your web flow cloud and you can have that you know mixed with your web flow project. So maybe they are like going this direction at least like marketing it because they want to market uh they want to advertise it to this type of people who people who would bring their web app to web flow cloud as well as their marketing they're like marketing website and this could make sense to them.

I mean if you bring your web app to web flow cloud you you already have your website on web flow probably so there is no point at saying oh don't use contentful use us yeah that's what I mean like unless they release this as a separate little thing there's just no point in using this unless you've got a web flow website already and you're already using web flow because no one's going to be like let's just buy web flow but only access the the API because that's what truly headless that's what a headless CMS is web flow you've got your terminology wrong here but if you've got the interface and all that then you might as well use it right weird weird approach so it's not a new tool but just to be clear this is not a new tool this isn't something new that's been released at all this has always been the case this is just a marketing tactic to try and maybe capture some developers who didn't know that you could do this with web flow Yeah, potentially.

And as far as I understand, it wasn't it wasn't um like mentioned as a new thing. By the way, Corey, if you're looking at if you're watching this, I saw this misalignment that this bothers me. You see you see it, right? I What are you talking about? This misalignment. What misalignment? This is not aligned. I don't I I see it as misaligned. Am I crazy? You're crazy. What are you talking about? Crazy. No, it's not misaligned. It's not. I'm sorry. It looks misaligned. Okay. Come on. Let's wrap up. We got three minutes. Let's go. Let's go. Let's go. There's anything else? No. All I wanted to say I I I got distracted because I thought this is misaligned.

All I want to say, they this was not about a new tool. This was just saying that headless CMS creates more work. Basically saying, hey, don't use headless CMS. That's it. Or be careful. Be careful, you know, be careful what you are getting yourself into. It's just a lot of uh work. But yeah, that that was it. So, let's see if you can pick up on this then. So, well, that's all we have time for. Make sure to follow us on Instagram and YouTube @ commandai show pretty much across all the platforms and join us next week for all of the news in AI design and dev. I've been Sam and I've been Kavarza. Keep on vibing.

Peace out. Yeah,