Are WordPress Themes Dead + Best AI Workflows ft. Matt Medeiros

Brian Coords (00:00)
Welcome back to webmasters FM. Today I have my first, I think repeat guest, Matt Medeiros, the pod father of WordPress, which he does not like me calling him that. welcome Matt.
Matt Medeiros (00:11)
Brian, it’s great to be back on your show. Honored to have the title of First Repeat Offender.
Brian Coords (00:17)
Yeah, I mean, it’s a pretty low bar. I’ve only had a handful of people. I kind of forget that I have this podcast and then a month goes by and I think, man, I really want to talk to somebody. But there’s been lot of interesting things to talk about and I think you’re the person to talk about them.
Matt Medeiros (00:32)
I’m delighted to be here and have those conversations.
Brian Coords (00:36)
for context, we’re Wednesday afternoon and you just released a video, was that this afternoon or was that yesterday? I’ve already like, okay.
Matt Medeiros (00:46)
That was yesterday where
I did a video on, it’s first time I’ve ever done a video where I read a blog post in like a sort of reaction video style of Kevin Geary’s blog post, which we’ll go into. But coincidentally, I also did another video which was like 48 hours before that about what I want in a WordPress theme, which,
Brian Coords (01:14)
Mm-hmm.
Matt Medeiros (01:16)
kind of jive pretty well with some of the concerns that Kevin had and I broke down in the video.
Brian Coords (01:22)
Yeah, I’m kind of curious to hear. So I, I sort of glanced over the article when it came out. I guess for context, I mean, I consider Kevin like a friend, but I also, he, him and I will definitely butt heads on like approaches to things or well, philosophical differences, but he’s a person I kind of always appreciate talking to. Cause I feel like I’ll get something out of it. I feel like he pushes, you know, ideas forward and I think that’s helpful. but I, I was just super busy.
This week was just like one of those weeks and I didn’t have time to like fully read his article. Like I kind of glanced it and stuff. So then I watched your video, which was actually perfect because you did, you read it. And then it was kind of like a, like a prime primogen video or something where they like, like a live streamers who love to like read blogs and react to it. I don’t know. It worked really well. So I’m recommending it to people. Uh, especially if you don’t like to read and you just want to have somebody read it to you, but also give it a little flavor. Um, what was your takeaway from his post?
Matt Medeiros (02:14)
Well,
and for some background, by the way, dear listener, Brian also held my feet to the fire in the WP Minute Slack channel and said, 37 minutes? And also complained about ads and I said, pony up for the YouTube premium and it wouldn’t be that bad. I agree with you and I think Kevin would agree with you about sort of how he approaches a lot of these topics.
Brian Coords (02:25)
It’s true.
Matt Medeiros (02:39)
As I say in my video and as I’ve said to him and I’ve said countless times, like Kevin should be defending CSS at the internet layer of technology. And I think he’s found himself in the WordPress ecosystem and is shouting at everyone saying it should be done this way, but nobody asked for it that way. But it’s perfectly fine from a product perspective.
Brian Coords (02:48)
Mm-hmm.
Matt Medeiros (03:02)
This is how WordPress has actually thrived for the last 20 odd years is here’s this core of WordPress, here’s this application of WordPress, here’s how it’s defined, and then hey, it doesn’t do exactly what I want, so a product team comes in and builds on top of that. Whether that’s a new theme, a tiny little plugin, or a gigantic ecosystem.
like an Elementor that comes in layers on top of WordPress. And Kevin’s doing that with, or will be doing that with, etch when it’s fully released. I have a lot of agreements in his blog post, but the biggest thing that I disagree with is Kevin’s perceived ownership of etch being the thing that solves
all of WordPress because it doesn’t. Number one, because it’s a paid product that has to be purchased to solve that pain point. And again, nothing wrong with that. But he is shining the light through some of the holes in WordPress, of course. And again, I’m not a developer. He’s very developer focused, but he’s shining the light through those holes to say WordPress needs to catch up to operate like this.
I can’t even say if WordPress should operate like he’s defining or should be built like he’s defining, but certainly we want WordPress to keep innovating. And he’s innovating for sure, but it doesn’t solve WordPress for the greater user base. And then I would also finish on saying he and a lot of the folks that love what he’s building just see WordPress as a tool for professionals.
who are developers, agency owners, and their main goal with WordPress is to build sites for clients. And that’s how they perceive WordPress, but WordPress is not that. Unless I’m totally missing the point, which could be valid, but WordPress is for the person who doesn’t know anything technical, who just wants to put words onto the internet. Now you can argue that that is,
Still difficult for that beginner person, but that is to me the essence of WordPress. I want to put words on in the internet. Might be a blog, might be a small business website, but that’s its intent. Its intent was never to build an agency tool for professional web developers, but I can also understand how that gets muddied as time marches on.
That’s my long sort of definition of what happened with that blog post.
Brian Coords (05:35)
Yeah. And I think, I mean, there’s, there’s so many different routes to go. think there’s like the open source versus selling proprietary software and the way people tend to treat WordPress. Like it’s something they paid for and they can get a little demanding on what they think it should or shouldn’t do. I think that’s one avenue. think the other avenue is like, who is WordPress for that classic conversation. And I think that’s the part where I tripped up the most in the post where the thing is with Kevin.
I probably agree with 80 % of what he wrote. And I think there are things that many of us in the block theme world have been talking about for a long time. There’s, know, we wanted a consistent design token system underneath that applied to all themes and all themes had to use the same and things were more transferable. Like we wanted a lot of these things, you know, a lot of the pain points he says are the same, but I think the thing that I tricked up on is
is let’s just take the example of when I change my theme in WordPress, I lose access to a lot of this stuff. And it’s this idea that someone’s going to go in and change the theme to their site to get a different style. Now, the person who’s going to do that is not the same person who’s going to buy etch and write code to build a custom theme for a client. Like I think those are two separate people. So I think the, the entire problem statement of everything’s locked into your theme and you can’t change themes and it’s, you know,
everything is sort of bundled and people get trapped is a different, completely different audience. And I don’t think handing that person etch is going to solve their pain point at all because I don’t think they’re the same person. So I think I struggle like who are we talking to? Who are these problems for, you know, even to begin with.
Matt Medeiros (07:16)
Yeah, I mean, I think. He even comments, I believe, on my video and says exactly that, like etches for the developer who’s going to, you know, build this out and then give the customer their website and the customer never has to think about it. Well, that’s fine if it’s a. Commercial situation like you’re selling a site to somebody and they’ve built it, but they’ll still need edge.
The biggest issue is, as he’s rightfully highlighted, is how do we get WordPress at its core to operate in a sane way that makes onboarding people to put their words or their website on the internet in a quick and easy fashion. I said, as just like a power user of WordPress and somebody who also sold themes and plugins many, many, years ago,
Like yeah, I wanna be able to build my site. And I think you and I talked about this, 2014, 20, or excuse me, 2024 theme, 2023 theme, I forget when this happened, but it was like, oh, if I build this out, I should be able to just switch themes and have those blocks there, but they’re not, they just disappear. And that is, as a power user, if somebody who just wants to get stuff done with WordPress, yeah, that’s the kind of ease of.
Brian Coords (08:28)
Hmm
Matt Medeiros (08:39)
ease of use I’m also looking for. I’m not looking to debate. Like the code that’s used to do that or like, you know, the JavaScript or is should it be CSS? Should it be HTML? Like, no, man, I just I just want that to transition over and I can get in. I can break down like my thoughts on like what I expect from a theme, but. Yeah, like as a power user, I agree with a lot of that easy use stuff that he’s that he’s he’s arguing as well.
Brian Coords (08:46)
Yeah, yeah.
Yeah. I mean, I think this is a debate that at least in the developer side has been around for years and years, which is how much do you bundle into the theme and how much do you separate into plugins like themes for design plugins are for functionality and you shouldn’t combine the two. And I think one example he had was, what if you have a portfolio post type or something, right? They put that in the, you put that in the theme and it causes all this problem and like.
Yes and no, because we would do that kind of stuff. Okay, well let’s make a separate plugin just for their portfolio stuff. And it’s like, well, I still got to design it. I still got to make sure it matches with the theme and the, and I still got to have patterns for it. So then I’m just kind of doing things in two places. And I think this separation of concerns that I is a really big issue for a lot of people. think it’s like an ideal thing that we all talk about, like my content and my design should be completely separate and I should be able to swap this and do this. And it’s just.
I don’t think the real world ever functions that way. I don’t think that’s ever been, I don’t think there’s any tool that does that. I don’t think there’s any system that’s really solved that. And I think it’s becomes this sort of this crutch. On the flip side, I do think the block editor was supposed to promise that a little bit more. I put in my gravity forms block and I put in my WooCommerce blocks and I can change my theme and the blocks are there and they work with everything. And it hasn’t lived up to that yet. But I think…
Comparing it to where we’ve been, it’s such a huge improvement that I think people underestimate it. And I think that whether the code is in the theme or in the plugin, whether I can’t change themes or I can, but I can’t turn off your page builder either way, it’s like to me, that’s all the same thing. The fact that you could in your site go into your 2024 theme with all your content there, they have like 10 style variations that you could switch, right? That would do exactly what you want, restyle everything.
but keep all the structure the same. You can change every single one of those in your block theme if you wanted to. I think the functions are there, but if we wanna argue about, that shouldn’t have been here, that should have been there, I think like, yeah, that’s fine, but like, know, fundamental sins of Gutenberg, I think is a little extreme.
Matt Medeiros (11:05)
Yeah, I mean, I think as I mentioned in the video. I mean, I look at that and I think I think if we were sitting at a word camp at an after party, sitting at the bar with Kevin and we said, yeah, you think 20 years that was all done intentionally, he would kind of spark a little bit and like, you know, tell us like, yeah, I’m kind of embellishing that, you know, whatever, like I get it. We all well.
Brian Coords (11:24)
Okay.
Matt Medeiros (11:30)
I don’t want to say we all do it, but we all do it in like the marketing and promotional world. Like it wasn’t, I don’t think it was done intentionally. Open source is messy. I’ve always like related it to local government. Pothole hasn’t got fixed for six months. What’s going on? Why is it fixed? The road collapsed finally. Okay, I guess we’ll finally go out and fix it. I feel like that’s a lot of open source. But let me ask you this question.
in terms of like that, what I’ll call data port, well, I’ll call it data portability, but it’s probably not the right word. Do you think it’s as easy? What if WordPress had custom post types, custom fields by default, and a really usable query block to build out those pages? it that, is that the 80 % to be done?
Brian Coords (12:01)
Mm-hmm.
Matt Medeiros (12:22)
is to have like an interface for people to duplicate what they already do with posts because posts and pages travel no matter which theme you switch to, they’re always gonna continue with you. If the basic user could just say, I want a portfolio custom post type, and they could just like copy paste the posts item in a very, very easy fashion.
Brian Coords (12:40)
Mm-hmm.
Matt Medeiros (12:46)
Is that what we need to get done and coupled with a query block that like could like build it out in the front end? Do think that’s like the 80 % we need in core WordPress for most people to have data portability?
Brian Coords (12:59)
What’s really hard about these conversations is every day, not every day, every week I sit down and think I’m going to write a top three priorities in WordPress, uh, post, it just depends on what I’ve been doing the last two weeks to what I think are the things that would be the most impactful. Um, obviously I would say that getting custom post types and custom fields into core would have been a huge, uh, a huge game changer and maybe not 80%, but close to that 80 % mark.
That’s why I worked on that project with wordpress.com to offer a way to do that. Cause I really believed in it. And I really think it’s that important. Um, you know, that, that, that is a thing that I truly believe in custom fields, custom post types, content modeling, get all that in core. Yes, that’ll change the thing. The other piece too, is that the other side of that is I think the design system stuff needs to move to core. So I think I said earlier, like if you have 20, 24 theme and you, it comes with like, I think a dozen style variations, you can.
change style variations without breaking your theme. It changes the colors, the fonts, the spacing. I think anything beyond that, you’re not gonna be able to change it, right? I can’t say like change all my, you know, three columns to be four columns and add extra content. Like layout changes just, never gonna be easy. They’re always gonna be manual and no one’s ever gonna solve that. think that’s, people bring that up and it’s a red herring. But yeah, so.
the content modeling stuff in core. And then also a lot of that stuff that’s inside themes, I do agree should be in core, your design system, all that sort of stuff should be hard set in core and people would be mad, but I think it would make it a lot easier. And then you could switch themes and cause I, cause I guess the example in the blog posts that I disagree with the most was this idea that if you like Brian Gardner’s powder theme and you like Mike McAllister’s Ollie theme, and you have a pattern in one theme, you can’t bring that pattern over to the other theme, but that’s just not true. You can, you just won’t get,
the same fonts and spacing and all of that sort of stuff. But I don’t think that’s a huge issue. You can take anything in any block editor theme and copy it. And all of the blocks are going to come over. You’re just not going to get that quite the same style stuff. And I know that for Kevin style is the most important thing, the CSS and all that stuff. so that’s why if, if we all just use the same CSS names and stuff, it would make it a lot easier. Sure. But like not a high priority to me. I would probably agree with you that the custom
content stuff I think is like the last missing piece in Core.
Matt Medeiros (15:23)
Yeah, and I just bring it up because I think when I when I read a Kevin Geary piece and even like try to follow, you know, other other folks who are building themes and all this other functionality, it’s just like I have to take a step back again as just like a non developer power user businessy kind of person and just be like, well, how far off are we like once the dust settles on both sides and both
you know, everyone with their particulars and their debate, like, how far off are we? Now, I know the challenge is how do we execute on it? But at the end of the day, like, I’m always just trying to find, like, what’s the least amount of effort we need to do to make the biggest impact? And I know it’s not that easy in open source or specifically for WordPress, but I try to look at it in that fashion, like, we’re not that far off. So is it a good thing that WordPress?
Like, do we just sit back and like, OK, Kevin has a solution, but it’s a solution for his users and that that avatar. And do we just not solve it and have people come up with these unique solutions that solve it for their particular avatar? mean, I can sit here and debate debate that all day long, but I just think that we’re always we’re never like that far away. At least. It doesn’t seem like that far away, you know, to me in the video that I put together about.
like what I want out of a theme, because this got a little bit, I got some feedback on that as people are still like, well, you’re asking for it and like the entire functionality of a theme, like all this functionality in my world at WP Minute, like I run what I would consider a proper media site. So I need things like multiple content layouts, featured content, podcasts, videos, a membership sign up page.
Courses coming soon newsletter sign like I need all these things I don’t need the theme to do the function But what I would pay for is a theme and I’ve said this to Mike a million times with with like Ollie to support Like pick a handful of major plugins like lifter LMS like gravity forms like woo like whatever and and think of those styles
And think about styling those pages for those particular products and charge me between 500 to $1,000. And I’m in. Now, maybe the business case isn’t there for scale, but I don’t know. I think that there’s something there for a theme author to build out a theme that works with these
Brian Coords (17:43)
Mm-hmm.
Matt Medeiros (18:02)
very popular plugins. And these plugin companies, as somebody who works at Gravity Forms, would love to partner with somebody like that. So yeah, maybe the days of a $59 single site license is gone, and this is what we need to move towards, a more comprehensive, larger ticket price theme that partners with other plugins to build something better for their customers. That’s just my own pie in the sky.
Brian Coords (18:09)
Mm-hmm.
Yeah, I have a few thoughts on that. mean, number one, obviously I’m, I, I’ve been going to Mike about Oli and woo compatibility pretty regularly. Cause I’ve been using it on demo sites to try to figure it out. And he always says to me, it’s not worth it. I don’t want to touch it, you know, cause it’s complicated. There’s a lot of templates, a lot of layouts. then, you know, a lot of times even inside WooCommerce, every problem you have is,
Matt Medeiros (18:45)
Yeah, it’s fair.
Brian Coords (18:53)
is our problem. could be your hosting company. No, it’s WooCommerce broke this or the theme did something or a plugin extension. No, WooCommerce broke that’s because that’s just the thing, the name that they’re attached to. And I can understand why if you’re Mike and you’re Ollie and you’re telling people I have WooCommerce theme support, the minute anything visual is off, he’s going to be there, you know, support person. So understand that. And also I will say I’ve had two experiences with this. Number one is I used to be the maintainer of a large open source theme.
called under strap. don’t know if you know this one, but it was very popular for a long time. ⁓ yeah, it’s, was bootstrap mixed with, you know, very classic old school WordPress. And we had WooCommerce templates and every, you know, few months they’d update their templates. We’d have to go through and do all the testing and push those into the parent theme and then all the stuff. And it was really hard because, you know, they want to add some functionality to their templates. We have to make sure that works. Then every developer who’s changed it in some way now has to make sure it works. And it’s hard.
Matt Medeiros (19:23)
yeah. Yep.
Brian Coords (19:49)
And now inside WooCommerce, we’ve been working really hard on a block theme that’s tailored to WooCommerce and to a lot of the blocks that are, you know, a lot of the screens that used to be short codes. Now they’re all block based. And it’s the same exact problem, which is, well, how do we give you a template that’s in blocks? And now you have complete freedom to change it, but now you’ve changed it and we want to add a new feature. What do we do? Break your site and give you the new feature? Don’t give you the new feature. You never find out about it.
try to put it where we think in your blocks, makes the most sense, you know? So every time you give that extra freedom, it just makes everything harder in the future because now every time we want to do something new, you know, we have to not break your site, which is the number one thing that WordPress cannot do. It cannot break somebody’s production site, no matter what version they’re on, that sort of thing. So I think these are just like really hard problems to solve. I think the block editor has made some improvements and I think people should compare it to what…
It was like in the classic theme era. And I think it’s easy for a lot of people nowadays that have come into WordPress through page builders. And to them, it’s like, well, I’ve been paying Elementor, you know, 500 bucks a year and they solve all these problems for me. And now I want to not pay them, but I still want all these problems solved for me. ⁓ but they’ve never been around when we have to do all this stuff by hand and what that felt like. And so it’s definitely easy to show up towards the end and say like, you did it wrong for 20 years. Here’s the right way to do it.
Matt Medeiros (21:00)
Right. Right.
Brian Coords (21:10)
because you have all the lessons we all learned the hard way. But in 10 years, you’ll be that person too. And everybody will see what you did and say like, can’t believe you did it this way. Why didn’t you know about this and that? And it’s just a cycle. It’s open source.
Matt Medeiros (21:24)
Yeah, and know, and that’s a total line to tow the line from from Kevin a little bit is like, which I also agree with is how much time do we have left? How much time do we have left? Because anybody like people can come to me like Kevin with with their etch and their Elementor thing and even the concerns like we talk about with Gravity Forms is
how much longer do websites and WordPress exist like they are today? And we gotta make a move. We have to make a move and we being like WordPress and the core contributors and the innovators behind the software, I am not that person. I am certainly here to provide feedback and to hold and prop up the successes of WordPress and the use cases of WordPress. But.
I wanna see it survive, so like how much longer? Obviously we’re not gonna answer that on this episode, but how much time is left? And I think that the good that comes out of these debates, whether you agree with Kevin or not, or like his approach or not, hopefully it spurs some innovation to happen faster. And hopefully some of this stuff trickles into the core software of WordPress because it is needed.
it is needed to survive in the future against all of the competition and of course, things like AI. AI in fact, might be a perfect use case. And it’s actually given me an idea unless it’s too far fetched is, yeah, if I transition that all the block to Brian Gardner’s block and I lose all that styling, could AI solve that?
Like is that an opportunity right there for AI to evaluate the styling from the old theme and then see the styling on the new theme and go, boom, let me apply the styling and maybe it doesn’t get it right, but if it’s 90 % is a few other things I need to adjust. Maybe like, there’s a good use case for AI. Anyway, I also urge that something has to happen for sure if we wanna keep thriving as I like to keep repeating that phrase.
Brian Coords (23:23)
Yeah. I mean, I, I all agree with your first point, which is that I’m always kind of glad when, Kevin brings these conversations up. even have, we’ll have situations, like working at woo. We’ll have situations where people in the community will kind of be a little aggressive. Cause I don’t know if you know this, but people in open source can be a little mean to you. especially when you work at a big company. but a lot of times you think you see that and you’re like, you know, I don’t really agree with your tone, but your substance is true. And I want more people to vocalize these things. Cause it doesn’t.
It helps, you know, cause like I said, I probably agree with, you know, more than half of what Kevin’s laying out. think that those are definite places to make some wins. But then like you said, there’s just this thing of AI sitting over your shoulder where you kind of feel like, are we pushing deck chairs around on the Titanic worried about, you know, user interface problems that honestly might not be problems in the future. I, I, you know, the last time we talked, I looked it up was January.
for when we did this podcast before, I think that was the last time we talked about AI and its progress. And so I feel like this is a good opportunity to check back in and see, because I think that was when you had made the Pulse WP, right? Probably site and you were burning tokens and stuff in the last six months. I, I didn’t understand what people meant when they talked about this. And then I tried Claude code with an AI, with an API key.
Matt Medeiros (24:31)
Mm-hmm, yep.
I’m still burning tokens, Brian.
Brian Coords (24:46)
and it shows you your API key dollar sign, just dropping and dropping. And so I’m here for advice because I’m starting to get into, I’m starting to get into UI building with AI. Like I’ve used it for actual code pretty heavily since even like the co-pilot days. And I I’m enjoying it all.
I’m trying to do that next thing, is I’m going to build my own micro SaaS app for myself to solve my own use case and blah, blah, blah, not going very well. Have you seen improvements since January or do you feel like the tools are the same way they were six months ago?
Matt Medeiros (25:19)
So I have seen improvements with, quad code has done some tremendous things. And by tremendous things, as a non-developer, now full ownership on that Vibe Coder title, quad code gets things right.
90 % of the time on the first prompt. And if there are errors that creep up, it fixes them with a follow-up prompt. For example, when I started, everything I, even back dating back to January, a lot of the projects that I start, and I still kind of do this today, is I start with Bolt. And I start with Bolt simply because it’s just in the web browser.
I don’t need to set it up locally, just throw it in the web browser. It’s fast and it allows me to build that like really, really early stage idea. where I still haven’t seen AI win, again, non-developer, let’s remember that, is or improve, is front-end design. No matter what I try, some people say, make sure you tell it to use the most cutting edge version of Tailwind and
Brian Coords (26:04)
Mm-hmm.
Mm-hmm.
Matt Medeiros (26:32)
You know, I tried Sketch by Google and all this other stuff and it’s just vanilla tailwind CSS styling. And that really has me thinking critically. It’s like, why isn’t front-end design? Like, if AI is so powerful, why can’t it come up with, give me 12 iterations of this screen that I’m trying to create with this app. Give me 12 iterations.
Brian Coords (26:40)
Yes.
Mm-hmm.
Matt Medeiros (27:00)
with this kind of look and feel and it can’t do it. And I’m just sitting back going, why can it do all this other really complex coding stuff but not this front end design? Not to say it’s not complex, but why couldn’t it just give me something else? So it certainly hasn’t won there yet. Spend some time in Bolt and then I, Pulse went to Replet because at the time I was just like, I want an all encompassing environment. Has Replet improved since January?
Brian Coords (27:20)
Okay.
Matt Medeiros (27:28)
Yes, not as much as I’ve accomplished with Cloud Code and other projects. Replet will still some, from time to time, you’ll ask it to do something and it will just rewrite a whole thing on your app, like still today, no matter what you try to do. And that’s the gap that I’ve seen really close in the last month or so that I’ve been spending a lot more time in Cloud Code.
is I don’t ever fall into that death cycle of vibe coding where it’s like, what happened? Let’s revert. Let’s go back. No, you didn’t do this right. And again, I’m just speaking naturally. I’m not doing any tricks. Maybe I should. But it’s just when I run into an issue in Replet, it takes me longer to get myself out of that hole.
Brian Coords (27:56)
Yeah.
Matt Medeiros (28:15)
than I ever land in with Claude Code on some other standalone projects that I’ve been working on. So that’s a long way of getting to, Claude Code has been really great. As a Vibe Coder, I’m less worried about taking on these projects and I feel more confident building it out on my local machine, deploying it to Netlify using SuperBase as a backend. I feel more confident using Claude Code. And the reason why I had gone with Replet was because I
I didn’t have that confidence. Like I felt like I had like an emergency switch I could hit with Replit and it would take time and tokens and money to dig myself out of some big feature. But now I feel more confident with Claude Code and I plan on actually pulling Pulse off of Replit and seeing what it can do if I come in on the backend with Claude Code and try to enhance it. So yes, from.
Brian Coords (28:45)
Mm-hmm.
Matt Medeiros (29:07)
month one to month seven and a half, I’ve seen definitely some improvements for the VibeCoder like myself.
Brian Coords (29:15)
Okay. I mean, that’s interesting because it does feel like maybe not the last six months, but the last three months for sure. Models have kind of stalled a little bit. It feels like there was gains gains and now we’ve kind of we’ve hit a bit of a plateau and it’s a plateau that, you know, is still miles above. Like I said, like I cursor for me is the one I’ll go to the most. I, my wife,
Uh, sell some stuff and I thought I’ll set up a woo store to kind of get a chance to play with it a bit. And, uh, and I wanted to make some custom post type type stuff to help her manage some stuff and everything. And I opened up cursor and a local environment and I tell it what I want and it’s great. I don’t have to do anything. I give it the WordPress kind of component react component library. So it looks all pretty good. It still wants to keep throwing CSS at it and like hand styling everything. And I keep just saying like, no, there’s a component library. Just use it.
and it, and it’s pretty good. I wanted to do the other side of it, which I think people do on like replator bolts, which is, is not, I’m going to just hack a little feature on WordPress, make a little plugin or something, but I’m going to make something from scratch. And the concept I had was like, I am unhappy with all the workout apps that I’ve gone through and I know what I want to do. And all I want is like that timer. That’s like, do this movement for this much time rest for this much. Like I just want the timer and the rounds counting done for me.
Maybe track the weights or something like that. That’s all I want. And I thought that seems like the perfect option and I’ve been building it and it’s exactly what you described, which is I asked for one tiny feature and then suddenly there’s entire screen of things I didn’t ask for. This other feature got removed. This other thing’s not working. It all looks ugly. Super ugly. I have to go in manually and just like get rid of like the, keeps adding borders to like everything. Everything is just like this.
Matt Medeiros (31:06)
Yes.
Yeah.
Brian Coords (31:07)
like it looks like 1990s software. It’s wild and I didn’t, I’m like hitting that point where I this has to be me. Like should I be, cause I used to like experimenting in the code and it’s like, I’m having an idea. I don’t really know. I’ll build it and I’ll flesh it out and I’ll keep iterating on it and stuff. But what I’m finding with this is like, if you don’t have a clear idea of what you’re asking for, like it’s not, it’s going to go really far down a road you didn’t.
want to and I’m like, what am I doing wrong that I can’t get it to just just do the one little thing I asked for and stop?
Matt Medeiros (31:42)
Yeah, I’ve noticed. So one of the like little hacks I have with Bolt is I’ve actually found Bolt to do halfway decent UI and. Gives me the best variation. Of of ideas, so like. If I asked it. In six different prompt windows to make me.
Like I’ve been working on this content calendar app. And if I asked it to just draw a calendar in the middle of the screen, and like that’s where I wanna start, because that’s like the most common block of UI that people are gonna work with. That’s how I started with Bolt was, I forget what the prompts were, but it’s just like come up with a modern looking, clean, minimalistic, know, notion like calendar view, blah, blah, like do this thing.
Brian Coords (32:21)
Mm-hmm.
Matt Medeiros (32:31)
six times in Bolt is fast, fast and efficient, just not for Mother Earth, but fast and efficient in a browser. And then it gave me the the best gap in between like all six variants. And then I would continue on the variant that I liked and to expand upon that idea a little bit more. And then once I get to a point where I’m like, satisfied with how it’s working, how I look and feel, then I pull it down locally.
Brian Coords (32:37)
Hmm.
Mm-hmm.
Matt Medeiros (32:59)
I still use cursor, but largely just for the file management, ease of file management to just see the directory I’m in and all the files. And then I just load up Claude code there. Bolt makes it easy to deploy to Netlify and to start with SuperBase. So that’s why I start there. It kind of just lets me quickly integrate the two primary keys, which are the hosting and the database.
And then I’ll pull that down and do the rest locally when I think I’m ready to like continue with this idea. But still from like from there, it Claude code isn’t really enhancing the UI for me. What I do like what Claude Coase does versus Replet is planning, even though there’s like a planning feature for Replet Claude code does really well at planning and outlining like what tasks it’s going to complete and allows you to
interface with that plan more effectively. And two, like it’ll often draw me, it’ll draw me in the command line, like how it might approach doing a navigation item. Like if I’m asking it to show me like I’m working on user management, and I was just like, hey, come up with an idea that’s going to be a great way to display the users and like the edit icon, the delete button, like show me how you might like lay this out. And it’ll actually draw it out on the command line, you know.
Brian Coords (34:00)
Mm-hmm.
Matt Medeiros (34:19)
not ASCII art, but it’s just like, you know, it’s using like the equal sign to build the, the Navi. So it’s like, it’s like helping you visualize that in the command line. And I’m like, dude, this is great. You know? So anyway, the efficiency is there with Claude code. I’m definitely seeing over Replet these days.
Brian Coords (34:20)
Yeah.
Yeah, yeah, yeah.
huh.
Yeah, I’ve tried Claude code because I’ll use cursor and I never really seemed to hit limits with it. I don’t know if I use it enough, but I, I think I have the $20 a month or whatever plan. It seems to be fine. I don’t know if it’s slow or fast or I’m on a rate limit or anything, but generally it does what I want it to do well enough. Um, and whenever, when I’ve tried Claude code, I’ve gotten into these cycles where I just see the dollar signs disappearing.
And the, you’re right, we fixed it. This time it’ll work. And to be fair, think I was asking her something a little complicated, is timer-based. And that’s not easy for any developer when you want to have a timer that’s going and it changes the state and all this sort of stuff and everything. But I just keep fighting it and fighting it. what made me want to ask you is because we were talking about this, because I think at some point…
we’re gonna branch past the hobbyist vibe coder DIYer. I think right now people say like we’re in the command line phase where all the best tools are in the really command lines and chats and you have to be someone like you who even if you’re not a developer you kind of understand technology a lot better than the average person and you’re tech forward. But like fast forward a few years you were imagining an idea like my iPhone can generate apps for me and sort of stuff like that. Do you think the current trajectory is leading to that?
still after like, like, where we’ve been and where we’re going? Or does it feel like we’re kind of at this like, wall a little bit?
Matt Medeiros (36:08)
Yeah, mean, first of all, just I’m just theory crafting with this stuff. And I’m trying to think of a world where coding apps is as well, to be quite fair, as as ubiquitous as like building a WordPress website, you know, in the in the bubble that you and I live in. We all know that there’s there are people that can just waltz into the room and be like, I sold the site for 500 bucks, you know, and they’re like, I’m a web developer. Cool.
Brian Coords (36:25)
Yeah.
Matt Medeiros (36:34)
Fine. And that’s absolutely how a lot of people start. It’s how I started. So there’s nothing wrong with it. But what I’m thinking is like, what’s that next step? Like how? Life, how civilization changing is AI? And if it is, then the next step is to empower people to do things that.
Brian Coords (36:48)
Yeah.
Matt Medeiros (36:55)
that only developers in that white collar, more educated, affluent person has been doing for the last 20 years on the internet or 30 years on the internet. If YouTube made the common person a plumber because they watched a bunch of tutorials and all of a sudden now they’re fixing their leaky faucet or running plumbing throughout their house, will…
Brian Coords (37:09)
Mm-hmm.
Matt Medeiros (37:16)
These companies like Anthropic, OpenAI, Google, and Apple empower people to that same degree, and in fact, even more instantaneously. If you look at Replet today, one of the improvements that they’ve done is as you’re building something, you have an inspect button like you have in Chrome tools or your browser tools. You can click on an element.
And in the agent window, it’ll give you like a Gutenberg block of like adjust, change the font, change the font type, change the padding. Like it’ll give you all of those things so that the common person doesn’t have to think about CSS margins, padding font. It’s just it’s a little UI, right? It’s like running Gutenberg or Elementor inside your AI prompt. And I see that as that next step.
Brian Coords (37:50)
Hmm. Okay.
Yeah, yeah.
Matt Medeiros (38:10)
of, I don’t know, dumbing down the software building process. So my theory with Apple getting to Apple is. And again, this is just a theory is they have Swift. I’ve never developed in it, so I have no idea how it’s done. But I think if Apple were to be like. Hey, we’ve got this AI thing you you don’t need you don’t even you can create anything you can imagine.
Brian Coords (38:24)
Uh-huh.
Matt Medeiros (38:37)
as an app, well maybe they’ll put up some guardrails on an iPhone to say, imagine your own to-do app, imagine your own gallery app, imagine your own social sharing app. And there’ll be guardrails to this stuff, just like there’s a lawyer theme, a restaurant theme, a creative theme for WordPress, where the average person might just go, that’s pretty cool, I can make my own exercise app.
Brian Coords (38:57)
Mm-hmm.
Matt Medeiros (39:04)
Right, that’s I mean, that’s Apple Fitness. It ties into that whole ecosystem. And maybe you can just create some kind of software like big Lego blocks that you you can’t go under the hood and adjust. So I think that that’s where, you know, Apple would go if they were to ever unlock like an AI coding platform of their own. Right. Super, super dumbed down version of a replet. Right. Because that’s the Apple way. Right. Is like.
Brian Coords (39:29)
Yeah.
Yeah.
Matt Medeiros (39:33)
Here’s this super basic way that the masses can use. So when I bring this up and the real hardcore developers are like, that’s never gonna happen. Well, I mean, I don’t know, maybe it’s not, but that’s like the Apple MO is like, here’s a super dumbed down version, notes, know, to do the two, that’s like, that’s what they do. And I could see a world where Apple gives that kind of creative power to somebody and even say things like, never have to search for an app in our app store.
Brian Coords (39:37)
Yeah.
Yeah
Matt Medeiros (40:01)
hell, even maybe even sell your app in the app store could be a thing, right? So I think that that could still happen in my outlook.
Brian Coords (40:04)
Yeah, yeah.
Yeah, I kind of go back and forth because I do think like on the one hand, you’re right. Apple’s created these really like simplified mass appeal experiences that, know, if you’re a power user, you can step up to the next thing. But if you’re not, you know, it’s, uh, you know, like I think of their camera app or something like that, where you’re not a professional photographer, but man, your pictures are looking pretty good, you know, thanks to your Apple iPhone camera. Um, and then on the flip side,
Obviously Apple’s AI implementations have not been great so far, but it’s also just like, where are we going to, how far are we from getting there? And I think one of the spots that I wanted to hear your thoughts on is, well, one, think that a lot of these tools give you the blank canvas to make something with AI. And I think people get a lot of really good stuff out of that. Open up ChatGBT, nothing’s there. Now I have a whole research report written for me and that’s amazing and that’s great. And that’s where it’s really powerful.
On the flip side, so much of the world is dealing with legacy software. So when you look at AIs, when it’s integrated in these other tools, it’s kind of not that great. Like AI, know, Gemini is great, blank page. Gemini and Google Docs kind of sucks. Like Copilot is great in some situations. If you ever used it in office, it’s not great. Yet those are the places where we need it the most, right? I mean, I can spin up cool new ideas every week. A lot of people are doing it.
Where are those ideas? Everybody’s vibe code in these things. I don’t see any of them, right? I see a lot of people talking about what they’re doing, but you’re not seeing these things become long lasting tools on the flip side. I want to make improvements to existing stuff that is already part of my workflow. think that’s going to be really important and it’s just a harder problem to solve. And I think it needs to be solved. And I think WordPress is that’s the thing WordPress is dealing with right now. Cause there’s 10 different WordPress, start your site with AI tools out there.
but where’s the take your existing site and make it better with AI tools. I don’t see those yet.
Matt Medeiros (42:11)
Yeah, I I want to see my insurance prices go down. I want to see legal contracts get a lot more evaluated. That’s what I’m doing now all the time. Just a side note, I bought a new robot vacuum during Prime Day. And it’s got a fricking camera on it, voice activation, all this stuff. But you have to agree to the user agreement.
Brian Coords (42:27)
haha
Matt Medeiros (42:36)
And I just copy paste, threw it into, I think I just did Grok at the time and just said, evaluate this for me and tell me where I should be concerned as a consumer for my privacy. And it just like gave me all these things that this agreement has in it. Like it’s going to store your images of your house indefinitely. It doesn’t tell you that. And I’m like, OK, I’m not activating this. Right. So I’m like, those are the real wins that I want to see with AI is like really improving our life offline. But
Brian Coords (42:40)
Hehe.
Yeah.
Yeah.
Matt Medeiros (43:03)
The thing I keep going back to with a lot of this stuff is I can get apps off the ground, as of July, halfway through July 2025, you still, to your point, you still have to be able to get that business side of it, if you were to sell an app, you still have to do all that execution stuff. And…
Brian Coords (43:20)
Yeah.
Matt Medeiros (43:25)
I haven’t seen anyone solve anything major. Like haven’t seen anybody take a vibe as much as I watch these YouTube videos. I haven’t seen anyone really succeed on these $10,000 a day apps that you’re getting hyped on Twitter and YouTube. I’m running an experiment now because you you’ve challenged me many months ago. Like let me know when you earn a dollar. I did earn a dollar. I did let you know that somebody donated five bucks to the podcast thing that I did.
Brian Coords (43:38)
Yeah.
Matt Medeiros (43:50)
But I am running an experiment now, like can I build this app and build the marketing side of it with a WordPress website and build a business case for it and then just sell it? Like that’s the challenge I have that I’m running for myself right now is like can I build this thing to a functional app that somebody could say I can run with that. I’ll run with it, you’ve built it all for me including the marketing side of it and the business case.
Brian Coords (44:00)
Mm-hmm.
Matt Medeiros (44:17)
I’ll run with it the rest of the way. I’ll just buy it off of you for X amount of dollars. That’s the current experiment that I’m running right now. But yeah, like these apps still like native LLM apps. I saw Claude today say that now it’ll integrate with even more services, Notion, Google Docs, like all this stuff. But how is that integrating into my workflow yet? It’s not. It’s not.
Brian Coords (44:31)
Mm-hmm.
Yeah.
There’s so many situations where, no matter what options are available, I copy and paste and I go to whatever chat, GPT or Gemini, whatever one is closest to hand and just paste it in there and tell it to do what I want. And you think that would have been a problem solved a while ago, but it, you know, I, I feel like it’s part of my job in the role of like working in, a, like in the tech world, just like you, like I have to put in the time to learn this stuff. Like I think.
If you’re in our industry and you’re not, I don’t know what you’re doing. You should be learning this stuff. That said, if I were a normal person, there’s absolutely no way I would be doing this. I’d just pay 20 bucks a month to the Apple fitness app and move on with my life and not dedicate, you know, whatever, probably three hours, four hours so far to not have a functional app. I have to imagine we’re going to get there at some point in the future, but it’s.
Matt Medeiros (45:20)
Sure. Yeah.
Well, mean,
if you took my, sorry to interrupt, if you took my Apple theory, because I actually had this debate with Zemansky about this, if you took my Apple theory and applied it to OpenAI or Claude or Anthropic, like they would have to become, again, to dumb down the user experience and to just onboard people faster. Like if the software is to be disrupted as much as like some of the people say,
Like you should be able to just create the app on ChatGPT and it’s hosted right through ChatGPT. So they’d have to go buy like Versel, Netlify, or just freakin’ build their own infrastructure that hosts it, right? And it would have to be, I’ve got my ChatGPT 20 bucks a month bill, and by the way, it gets me five apps that it’ll host or whatever. But like, it can’t be,
Brian Coords (46:05)
Yeah.
Mm-hmm.
Matt Medeiros (46:28)
It won’t be disrupted if the current workflow is you can create anything on ChatGPT and then somebody goes and does it and they go, okay, now it’s just like WordPress. Like now what do I do with it? ⁓ you gotta go find a host, you gotta go find a backend, you gotta do this, you gotta do that. Like it just has to be effortless. Like now just host it. And they kinda have that with with Claude artifacts. And I think maybe ChatGPT has something where you can share like that visualization, but it’s certainly not a full-fledged app. So that’s how they would have to solve that for.
Brian Coords (46:37)
Mm-hmm. Yeah, yeah.
Yeah, yeah, yeah.
Yeah.
Matt Medeiros (46:56)
for you to be like, well, I got my chat GPT, 20 bucks a month, and I got some apps with it. That’s cool.
Brian Coords (47:02)
If you, yeah, like it’s funny that you say that about the hosting piece and stuff. Cause we just had a, like a public community call today with WooCommerce. And you have one person saying, you need to make it easier for me to just get it hosted and get the, I just want, I don’t want to think about hosting. want that fast SaaS. And then you have another person saying, don’t you even talk to me about hosting your open source. And I want to host it where I want it. And I don’t even want you. And it’s these two different sides of, you know, our, wonderful community.
And then I see both sides of it and have my own opinions. We have our own opinions internally. I think they’re both valid points, but I think solving that sort of like ease of use thing. But to bring it back to WordPress, which I think has traditionally not done a great job at ease of use and onboarding and stuff, it’s always been a roll up your sleeves and get it running type of a solution. And I think some hosting companies sort of solve that. But for the core product or the core, you know,
open source part of it. If you were like head of WordPress, you’re looking at AI, you’re seeing where things are going, all that sort of stuff. And you had keys to the kingdom, reference. what would you be, I’ll put it in the show notes. What would you be doing? Where would your focus be now that, you know, now that we’re a good few years into this, and I think we’re starting to see where things are going. What would you be doing for WordPress if you were in charge of that AI team?
Matt Medeiros (48:07)
Love that post, by the way.
Ooh, that’s a good one. It would certainly, on the essence of onboarding users to WordPress, it would be, my core mindset would still be around develop anything with WordPress. And I looked at the .com AI tool a few months ago or whatever it was. But how can we start?
from scratch to get people on board with building for WordPress. Let me go back to the Apple thing for just a second. Like I think that if they had some kind of AI app building tool, it would be a great way to onboard the full experience to Swift, right? And to become Apple developers, right? And build more apps and buy more app, Mac software. I said the same thing about WordPress and Tumblr. Man, wouldn’t it be great?
Brian Coords (49:11)
Mm-hmm.
Matt Medeiros (49:18)
if Tumblr was powered by WordPress and everybody had a mini WordPress site under the hood of Tumblr and then when they were ready to upgrade themselves, they could get a full-fledged WordPress site by just moving things over. And I think one of the things that made WordPress great was being able to build with it and learn with it. And I would keep that thread with AI. So give me an AI sandbox to build my WordPress anything.
with WordPress blocks and code and whatever, theories, and start educating and being able to build WordPress sites that way. And then starting with the playground, right? I can start with the playground, I can talk to the AI, can build these things, and then I can launch this project into any hosting provider that I want. Sure, give me a starred pressable account because it’s an automatic property, no problem.
but then allow me to download it just like I can do with Bolt. And really start leaning on, you can build anything with WordPress with AI.
From there, I would have to check myself on understanding, because one of the concerns I have is then what do you do after that? If you were to take that site and move it somewhere else, how do you extend AI into the WordPress experience? I think it’s easy to say, just have a little option that said, here’s your OpenAI key, and ⁓ plop it in.
Brian Coords (50:37)
Mm-hmm.
Matt Medeiros (50:39)
then you’re just favoring open AI as like, you know, the one, you know, the one AI system that you’d integrate to. So I would have to think longer and harder about how you continue with AI with WordPress. But I want to see a win for developing with word as easy as I can do it with Replet, with freaking JavaScript and Tailwind. I want to be able to do that.
Brian Coords (51:05)
Mm-hmm.
Matt Medeiros (51:07)
with a core WordPress theme and Gutenberg blocks. And in the future, like a decoupled environment, like something that’s, like I can just build a user registration system with WordPress. I can build a WordPress with just posts, no pages. I wanna build my own WordPress thing in the future, thanks to AI.
Brian Coords (51:29)
Yeah, I mean, I think that’s a good point. think obviously you struggle, you stumble on the hardest part, is the open source project can’t actually have AI in it. you know, and I, as, and I do think it would be amazing if the AI project had like, not even put your API key, but like one click opens your chat GPT account, know, like a, like a sign in with Facebook, but sign in with open AI type of thing. Like I think they could do that. Um, I don’t think that’s a hard problem to solve. Um,
Matt Medeiros (51:39)
Right, yeah.
Yeah. Yeah.
Brian Coords (51:58)
I think the other flip side of it is that I think we’re at that turning point where AI is going to start, like the screws are going to have to start getting turned on people on the heavy AI users. think we’re going to have to start seeing the prices catch up with the features. And I think that’s going to be an avenue where a lot of people are going to want to in the same way that, you know, set up your Stripe integration and the transaction person gets a little cut of every piece. I think we’re going to start seeing that with AI where it’s like.
don’t authenticate through AI, authenticate through GoDaddy’s AI, which is just open AI, but now we’re getting a little piece of it and we’re able to kind of rate, rate it and stuff. So I think that’s all going to be, it’s going to be another battleground for all that. But I definitely agree with you. And, I’m curious to see where, what goes next and what comes out. And, you know, I have my own thought. keep wanting to write a thing of like, here’s what I think should be the AI priorities in WordPress. And I just keep getting distracted, but,
Matt Medeiros (52:33)
Right. Yeah.
Brian Coords (52:53)
I think it’s something like you said, quick and easy wins, you know?
Matt Medeiros (52:56)
Yeah, I think one of the things about, and I think I’ve said this to you before, and since you’re a developer, you can tell me, when you look at, let’s say, SuperBase, Replit, because I think you were talking about the auth system on Twitter the other day, user authentication. That is, when you look at what holds a customer from a business side of it, if you build an app on Replit,
Brian Coords (53:12)
Hmm.
Matt Medeiros (53:20)
and it becomes successful, let’s say, you’re commercial endeavor and people are registering, signing up, using the Replet Auth system, that part of it, like you might be using open source JavaScript tools, Tailwind, all this stuff to build your app, but that Auth system is proprietary to Replet. No different than SuperBase, it’s a proprietary Auth system and it’s solving those more complicated components of an app. AWS has theirs, like so on and so forth.
Brian Coords (53:38)
Yeah.
Matt Medeiros (53:49)
So that’s like the cornerstone. Like I think WordPress has user authentication. Right, so if WordPress moved to that Replet-esque experience, something that we’ve all debated for a long time, is it an app, is it a website builder? I think with the help of AI, maybe it can be, loosely speaking, a Replet alternative if all things were equal. And now my auth system and my.
my database and the way that I store content and retrieve content and display content is all done in an open source WordPress way. To me, I’m there for that. I will build my app in WordPress versus Replet if that were the case. And I think you’d have to do that in order to make WordPress and open source survive. Otherwise, people just go, well, this other thing is easier and it just builds it.
So I’m just gonna use it.
Brian Coords (54:36)
I thought deeply about doing this workout timer on my WordPress site. was like, it would like, it like, I have the database, I have the rest of APIs, I have all this stuff. I build this, like this would be in my wheelhouse and everything. I ultimately was like, no, I want to try these other things, see what other people are experiencing. But, um, I kind of agree with you. It just feels like there’s a huge opportunity to say, we already have this thing. It’s already being hosted somewhere. You’re already paying for it. You’re already like.
Matt Medeiros (54:37)
That’s what I have.
Brian Coords (55:03)
Now you can start building those tools in here and it’s there and it leverages all that stuff. But I think it’s a long road from here to there. But hopefully we’ll see it.
Matt Medeiros (55:12)
One point I do want to make, and I’m kind of upset that it’s an hour in that I’m making this point, but when I try to build plugins for WordPress with AI, it is not as satisfying as like building a React slash Tailwind app because at least the way that I see it is the changes and the modifications and the features that I’m making are happening instantly.
Brian Coords (55:25)
Mm-hmm.
Matt Medeiros (55:36)
because of that, like I’m looking at, like I have half my window as my browser, I have the other half as cloud code, and literally as it’s doing it, I’m seeing those changes applying because of the nature of whatever, React, JavaScript, Tailwind, all these technologies. And when I do it in WordPress, it is a fricking struggle to be like, I gotta go to the, I gotta update the plugin, go to the page, hit refresh, okay, there’s the change. Did it work? Nope, go back.
Brian Coords (55:39)
Yeah.
Mm-hmm.
Matt Medeiros (56:04)
You know, and it is just not as exciting. And that is my worry for other idiots like me that come into this space. We’re like, I want to build my own thing. And my God, JavaScript is in React. I can just see it happening, man. It’s just right there. WordPress, I got to make a plug in. I got to refresh the page. I got to find it over here. I got to go to the app. It’s just not as satisfying. Can we make it more satisfying? You know, is the long term goal for me anyway, or what I would promote.
Brian Coords (56:05)
Yeah, yeah, yeah.
Yes.
No,
no, no, no, I fully agree with you. there is, but I, and I think that’s been a problem for all, like that’s always just been a problem with WordPress. It’s just that thing of like, and I also think like you’re generally solving weirder problems with WordPress. You’re like, you’re almost always solving two UIs at once, you know, like when you build a block, there’s like, how does it look on the front end of my site? How does it look when I interact with it? And that’s a
Matt Medeiros (56:41)
Sure. Yeah.
Brian Coords (56:58)
Almost everything in WordPress is like that. How does the settings page look? How does it change the behavior on the front end of my site? And it just adds so much complexity. Whereas like my workout app is just one screen. Everybody sees the same screen, that sort of thing. so I think some of it is a bit of like the, different types of problems that you’re solving. and some of it is the fact that it’s just this PHP site that you just have to reload every page constantly and all that stuff. But I think that at the end of the day, I think that you’re right. And to me, when I think of.
the WordPress site editor. And when I go into Gutenberg and have that black sidebar on the side, and I see my site inside the wheel, white bubble, it’s like to me, I’m like, we’re there. Like we’re so close to what replet looks like, what bull looks like, what every other interface looks like. Like here’s my site. Here’s a little area. Let me start talking to it. Like it just feels like it’s right there. And then maybe let me look at the code and give me some sort of like sandboxed code place where I can do it. But, but yeah.
Matt Medeiros (57:36)
Sure. Yeah.
Brian Coords (57:52)
there’s a lot of ground to cover, to even get there. And it has to kind of start with a really defined vision and a really defined audience of like, well, we can’t make it for everybody, but like, what’s the one problem we could make much easier to solve with AI?
We’ll see.
Matt Medeiros (58:06)
Would you pay 20 bucks a month to have a collection of what I call utility apps? if you had the desire to make more utility apps, like I have an app for podcastdownloader.cc, need to, oftentimes I need to just quickly search and download a podcast episode for various reasons. I have that. I have WPAPIexplorer.cc or .com.
Brian Coords (58:19)
Mm-hmm.
Yeah.
Matt Medeiros (58:30)
I have a new app that I just made, which I haven’t put on a URL yet, which is just an audio recorder. Sometimes I just need to quickly record audio and have some notes next to it. So I have like a little note section that I can just quickly look at the notes and as I’m recording the audio and then I just download the audio. Like I don’t want to boot up Descript. I don’t want to do like I don’t want to boot up Descript, have it try and do all that stuff, save it as a project. I just need to quickly create some audio and then upload it to Transistor.
Brian Coords (58:37)
Hmm.
Okay.
Matt Medeiros (58:58)
Like I have a range of utility apps that I’ve built that I actually keep using.
Brian Coords (59:02)
Wait,
wait, that last one, QuickTime doesn’t do that for you? Just like QuickTime, new audio recording? What’s wrong with that? What is, what? Why not QuickTime? I’m curious. It’s built into your Mac. How did…
Matt Medeiros (59:05)
No, no, no, no, no, no,
then it would have to be quick time and then also like, like if I, I’ve been recording episodes for a math report recently, right? And these are usually just like off the cuff thoughts, but now it’d be quick time and then I have to open up a notes app type of, no, I just, if I just go to,
yet to be revealed URL, I just have it right there and I can just create my notes for that episode and just hit record. It’s even faster than having open up QuickTime and then when I exit that note app, do you wanna save this? No, like I don’t wanna save it. I just wanna just get it out of here, right? So these are utility apps that improve my workflow. So the question is like, would you pay that 20 bucks a month to have a?
Brian Coords (59:36)
Mm-hmm.
Yeah.
Yeah.
Matt Medeiros (1:00:00)
Like if you had more utility apps like that exercise app, is it worth it?
Brian Coords (1:00:04)
I mean,
I think that’s the paradigm shift that we’re hitting is like the, I don’t really like the way, like the fitness app thing. Like there’s a million, I’m sure I could find one that’s exactly 99 % of what I want and pay 20 bucks a month for it. I’m sure that I could like, but it’s now hit the point where it feels easier to just make it exactly what I want. And I would pay 20 bucks a month to be able to do that with every single little idea that I have.
I think the flip side is like, has to get, it has to be faster to from idea to like final. And it’s still, I think it’s still hours of time to do it. And it has, we have to get to that place where it’s not hours, where it’s much faster to get from a, an idea. Because if it was, if I was just making another to-do list app, that is exactly like the to-do list app that’s built into my phone, you know, that’s not gonna, there’s not really a market for that. But when.
I want that weird thing. want that to do list app that does that one weird thing. You know, that’s that’s still the place where it’s like if we can get there, I think that’ll be pretty awesome. But I don’t think we’re I don’t think we’re there yet.
Matt Medeiros (1:01:09)
Fair enough.
Brian Coords (1:01:11)
All right, those were kind of my big questions. I mean, I had so many other big questions, but those are sort of my big ones. I think we got to do this sooner than six months before the next time that we talk. should people go? Which microsas are you promoting today?
Matt Medeiros (1:01:19)
Yeah, man.
Always down to talk about this stuff.
God, you can always find me at my day job at gravityforms.com and the WPminute.com is where I spend most of my extra free
Brian Coords (1:01:35)
Awesome. Thanks for stopping by.
Matt Medeiros (1:01:37)
Thanks, Brian.