Breakdown - A Gravity Forms Podcast

[00:00:00] Five, four, three, two, one.

Matt: It's been probably, what, like, eight or so months since you were on the podcast last. If I was a good podcaster, Jonathan, I'd have my show notes ready. Show notes, show notes. With the actual date that you appeared last. Um, you were building Hey, you know what's good for show notes is a little bit of AI.

Yeah, a little bit of AI, or a lot, or a lot a bit of AI. Let it do that. We're talking to Jonathan, uh, Williams about, uh, the stuff that he's been putting out, and, um, as we alluded to, it's, uh, as we were chatting before the show started, AI is really augmenting his experience. Do you remember back when you first came on the Breakdown [00:01:00] podcast, were you leveraging AI back then, or are you still doing it the old fashioned way?

Jonathan Williams: I mean, I, you know, I, I was doing both and still am doing both, depending on the circumstance. I'm using it for more things, I guess. In fact, one of the ways that, you know, you know, you look like you're officially an older person in technology, is it like you have to remind yourself there's a new thing that works better than the old thing?

And for some examples of that, I just mean that I, I continuously have to remind myself, you know, I will still find myself like opening a code editor when I have an idea, rather than starting a chat, which is just better, better all around. I've been thinking

Matt: about that too, like, I'm 43? Forty two? Forty three?

I don't even know. Uh, forty two or forty three. And I've, what I've realized is, Oh, we're living in the moment right now, when like the internet was first born, and we were younger, uh, at least I was younger, uh, back then, and it was like oh, this is [00:02:00] easy, like IP address, uh, dial up modems, no problem, grab it, stick it in the computer, connect to the internet, and you know what a chat room is, it, it became something where you were learning At a rapid pace, and you were just absorbing it, and you were the one ahead, and everybody else was behind.

And now I feel like I'm late to the AI game, and I feel like, wait a minute, I just realized I'm the old person in the room now. Like, there's people that have been using this for a couple of years that are far ahead, and they're younger than me. How did I miss this bus? And this is what it feels like. This is what it feels like.

And it's strange, and I'm trying my best to learn and adapt as fast as possible.

Jonathan Williams: What's also interesting to me is to, cause you know, we're all on some spectrum of this. Of this process where we're learning what these tools can do. And of course they can do more and more every day. So it's a moving target.

Um, but I, it's been, it's been interesting to me to see, uh, other [00:03:00] programmers, you know, especially those who were like really resistant to it at first, that some of whom I follow on X or Twitter, whatever we want to call it these days, but how many people were sort of really resisted to it at first. And then.

And then seeing some of those guys, you know, post their first, Oh God, tweet, where it's like, you know, tried cursor for the first time and gave it, you know, an actually decent prompt or tried any one of these tools. But I think part of the reason for that slow adoption amongst not just programmers, but a lot of people is that, you know, so many times since, like, when you were talking about that heyday of the internet, we, we were told that this next big thing was going to change everything.

And it didn't, you know, I'm thinking, I mean, I would put crypto is like the, the big one on there that was supposed to revolutionize, you know, every aspect of computing technology. If you talk to those blockchain guys for a couple of years, um, and, and this is completely aside from whatever you want to use cryptocurrency, but blockchain was like, it was all anybody could talk about.

Right. And I think web two had some of that and there were smaller instances, but. [00:04:00] But this time's different, I guess is, is the point, as anybody who has played with these tools figures out pretty rapidly this time is different.

Matt: Yeah. I, I, I wanna save that conversation for the, for the end of the episode because I, I think there are some roadblocks there, and I, and I want to get your perspective on the future of ai.

Um, but your website, ajar.com, you have a bunch of gravity forms, add-ons and other things that you're, you're building. What's the latest, uh, product that you've come out with? It seems like you're coming out with something new. Every week, somebody on my marketing team says, have you seen what Jonathan has built?

And I was like, yes, I follow him. I see it. I got to get him on the podcast to talk about it. Um, but what's the latest and greatest that you've built?

Jonathan Williams: Uh, the latest, uh, I, the latest plugin that I released wasn't actually for his Gravity Forms add on. It was my first add on for the block editor. And it was called, uh, Quick Link Pro, and, uh, what it lets you do is, uh, create and maintain a list of external links, which are then [00:05:00] automatically inserted into the block editor's standard link dialog.

So, you know, as you start typing in the block editor and the link dialog, it'll start recommending internal pages to link to. So, what happens now is I You know, the plugin gives you an admin page where you can add all the external links, whether they're affiliate links, or I link to Gravity Forms all the time, so I've got that in there.

And then those links, they appear just right there in that list of suggestions with the standard block editor suggestions.

Matt: Uh, on the website, you also have list user entries for Gravity Forms, loom for Gravity Forms, gist for Gravity Forms, product field for ShortCart and Gravity Forms, global search for Gravity Forms, which I believe is what I had you on for last time.

And, and a bunch of, a bunch of other, uh, Um, products, you know, what's the, what's the workflow like? Like it, you are pumping them out, um, but they don't seem like they're just tiny little features. They, they feel fairly robust. So there's some time and effort there. What's it look like to get these things off the ground?

Jonathan Williams: So for me, I start, you know, you start with a concept, [00:06:00] right? And you know, you could, you could look at like, um, list form pages for, for gravity forms, for instance, is one of my plugins. And that just, um, puts a, uh, a list of. Of the pages where your forms are embedded in your actual forms list. So, you know, when you're looking at your forms and gravity forms, there's a little site, a new column that will just tell you this form is embedded on these three pages and with links to them.

And, you know, so I started with that idea and I remember putting it in, you know, and I just start with the chat box. My favorite utility where I like to start everything is called typing mind. That's a paid sort of a multi chat agent. So you can chat with all the different models that you have API keys for.

And, uh, so I like to start in TypingMind and it works right in your browser. And, and then I've got, you know, uh, a preset prompt. Like, uh, I think TypingMind calls them, um, agents. But, uh, you know, a preset prompt to say something along the lines of, you know, You are an expert Gravity Forms developer. Um, you're familiar with the entire ecosystem of Gravity Forms products, and I even list like some of the, the [00:07:00] major other add on, um, companies out there like Gravity Whiz, Gravity Kit, um, Gravity Flow, I'm sure I'm forgetting some, um, and then, and then I give it like a bulleted list of these are things you do as a Gravity Forms developer, and some of those are like specific technical tips.

Don't do this, right? For instance, I'll tell it, don't ever use PHP shorthand syntax, because I just don't like it. Um, and I'll tell it, you know, make sure you use WordPress coding standards in your formatting. Um, and I'll tell it, you know, when you're working with Gravity Forms file uploads, remember that they're saved to JSON arrays, not serialized strings like you always think.

It's interesting, again, you know, when you work with these models, they, they sort of, they give bad ideas sometimes, and it's just really difficult to talk them out of, so you kind of have to say, Do this, don't do this as you as you go down those lists, so I start with that like this preset prompt. I start with the idea.

I go to typing mind. I usually start with either Claude 35 as you know the model that. Most people start with, with programming things, but [00:08:00] more and more I'll use, um, Oh one or now Oh three mini from, um, open AI whenever Claude gets stuck. And again, what I like about typing mind is I can bounce back and forth between those models in the same chat window really quickly.

Um, and so, yeah, the process is, and then we just go from there and then I'll, I'll basically, I usually will give it a fairly broad overview, not terribly detailed. And then with the understanding that that what it spits out at first will be in some ways better than I asked and in many ways worse. Um, and so that's the starting point.

And then from there, it's just, you know, rather than coding back and forth, you're just talking and iterating back and forth. Um, and then eventually, uh, when I get, you know, a certain amount of work done that looks to me to be pretty good. Then I, I dump it on a testing site and, um, pick it up in something like Cursor because of course Cursor AI is a full IDE so you can edit all of your plugin [00:09:00] files or WordPress installation files if you want to, um, you know, from a different environment.

So it's a weird kind of workflow, um, but it's, it's, it's sort of worked for me rather than, you know, there are people who do everything in Cursor which is great if it works for you. I like the, the free form of the, of the chat window more, especially when I'm trying to get like my head around what we're building and how it's going to work.

Matt: Yeah, that, that, I, I've spent some time over the last few months, certainly since the last time you and I talked, learning how to, uh, utilize these tools, uh, quite frankly, learning how to code outside of, uh, of WordPress with, uh, you know, React and other JavaScript frameworks and, um, You know, TypeScript and all this stuff, uh, that is a very, that's a very common sort of workflow of how I got started.

I was doing it in Claude one day and I was like, ah, let me just make this little thing that just does this thing for the download podcast episodes. And it like wrote the code and I was like, wait a minute, like, [00:10:00] oh, this works like this thing works. And then I start asking it more and more and as a total noob, like somebody who doesn't know how to, you know, Code outside of WordPress.

I'm like, wow, this is really powerful. And then I would do exactly what you did. I would paste it into, um, these other environments, but I started hitting those limits of using like native Claude or native chat GPT, where it's like, Oh, you're out of tokens come back tomorrow. Literally is what Claude would say.

And, um, that's when I discovered Cursor. So this is a long way of getting to my question. I hear a lot of people, you know, well, I hear. I see a lot of people on Twitter and other communities that I watch. Yeah, like you said, they all start in Cursor, or they're all starting in a native app. But, here's the rub.

I like how you're starting in more of a conversation, because for those of us that don't know what we're doing, um, you really want to flesh out that idea before you're, you know, applying that code. And I think that's super smart. You're bouncing [00:11:00] around LLMs, you're having a natural conversation. I think that's super smart.

Um, you know, and I, and I think that's a great way to, a great way to do it. Sorry, I said I had a question, but that was more of just a reference, thought reference. Uh, and I think, I just think you're doing a great job by approaching it that way.

Jonathan Williams: But that also brings up an issue that I don't think enough programmers, especially, or at least the ones that I, that I follow, um, have figured out yet, and that is How do I want to put this?

Um,

if you use the official app for a chat, GBT or for Claude or for anything else, and even if you access these models through cursor, you are in some ways getting a degraded experience versus accessing the API directly. Uh, now let me say what I mean by that. Um, so. It's not that I'm not saying that they're trying to make your experience worse, but [00:12:00] there's a couple of things going on.

Number one is that, you know, obviously, aside from the obvious, which is just the limits on some of those, um, on most of those official clients, right? Like Chad, GBT will eventually cut you off from the, from the higher models. And Claude does too. I think Claude is sort of more infamous than, of that than, than others are.

The performance you're getting is somewhat degraded, whether it's in, you know, the form of hard limits on how many times you can just, just ask a particular model version, uh, in a 24 hour period. But in products like Cursor, you know, they have their own magic behind the scenes that's trying to help you.

Um, you know, it's not, it's not just a straight connection to whatever LLM model you've picked in, in the cursor settings. They've got their own models in the back, and they've got some processing in the background, and these things are trying to help you. But, but, but anytime you're in a relationship like that, like the, when you're going through another party for, you know, for paying for the access to the AI that's behind it, You know, the company kind of has an incentive to [00:13:00] restrict your usage just because otherwise, you know, the costs will bloom.

Which is a long winded way of saying something like TypingMind or something like Bolt, which is a native, uh, Mac app that does something very similar, lets you access multiple models. You know, you gotta go to those sites and you gotta sign up for your own API key. But then, you're paying per use. You know, there's, there, you're never gonna get cut off unless you can't pay for the bill, you know, whatever usage that you're actually doing per, per use, per token.

And You know, it's worth it. If you've never tried that, I would just say to anybody developing anything out there, if you've never tried that, just, just try direct API access through one of those other apps. Um, because you will notice a difference. And that, that's another big reason why I start the way that I do with, you know, that, that software that lets me get direct API access to all the models.

So I get like a broad overview.

Matt: Yeah, it's very similar, right? If you're just somebody who uses WordPress, maybe you're not a power user, you don't consider yourself a developer. Yeah. Yeah. Um, you know, you have that default WordPress experience, [00:14:00] kind of vanilla, kind of middle of the road. But then, when you buy a theme, maybe say something like Cadence, certainly like Elementor, like, you're jumping leaps above, like, the core experience of WordPress, and it's made for that thing.

You want to build websites, you want to build webpages. Um, same could be said, like, Gravity Forms versus Contact Form 7. Contact Form 7, pretty basic contact form. You need more? You go to Gravity Forms, right? Specialized tool to help you develop this stuff. And I found that to be the case certainly for using Bolt, using Repl.

it, using, um, Cursor, you know, to get these things done.

Jonathan Williams: Repl. it's great. Repl. it is so good. God, I wish they supported. PHP in general, better there, you know, their agent, I've built, um, a bunch of standalone little tools and, you know, the, the way that their agent works, I think it prefers either what Python or, uh, react basically now it'll, it'll push you in one of those directions.

And I'm not a Python guy, you know, I mean, I'm, I'm, I'm, I'm [00:15:00] comfortable, um. You know, with PHP and MySQL and stuff like that on my own. But, um, but when I, you know, when I build those apps in Repl. it, it's just like you were talking about. I am the noob. I mean, I don't know the syntax. I guess you, you know, some things carry over between languages, even if you're not certain, but.

But it's very much that experience of, you know, I'm not going to get in here and edit this directly. Like, I'm going to have to just talk it out with the model.

Matt: So how do we approach the folks that are out there that are saying, look, AI isn't going to take your job. AI isn't going to, um, you know, build everything for you. I tend to agree with them now, but now that I'm sort of like, kind of knee deep in it, I can see a world where, hey guys, just like everything else we touch on the web.

Five years from now, it's going to be vastly different. So, okay, maybe not right now, but I can see a future where there is a sustainability coding bot that just follows behind you and says, okay, [00:16:00] you want me to make some patches to this product you made? You want me to make some feature updates to this thing that you made?

How about I scan for security bugs and it'll just continue to build it for you? Um, that's where I start to worry, like, all right. Low end and medium and middle tier coders, we might be in trouble, um, with this stuff. What are your thoughts around that?

Jonathan Williams: So, I think like everybody, I'm of like three or four minds on it.

I could see it go so many different ways. I'm not a doomer. I could start there. I do not think, you know, that this ends with, um, all programmers unemployed and, you know, humanity all transformed into paperclips. In fact, I find a lot of a doomerism to be so actively silly. I mean, I mean the worst of it. I'm not saying there's not going to be any, any new technology is going to have negative effects as well as positive ones.

It's history. Um, but yeah, there's, there's, there's just like this sort of.

They're not even really programmers, most of them, you know, they're not in here building stuff every day. They're just kind of sitting back [00:17:00] and telling us we're all about to die, um, which is helpful. But I also don't buy the straight up doomerism because we've never seen what happens when we separate intelligence from consciousness.

I mean, let's get deep here, right? But, but we talk about, we talk about artificial intelligence as though it's going to come with every, as though it's automatically going to come at some point with everything else that we associate with intelligence in humans, which is consciousness, uh, you know, feelings, the search for meaning, um, you know, in, innate desires that we are programmed to go out and find.

And there's just nothing that we have seen yet, caveat there, nothing we've seen yet. Suggests that that's true, at least in the way that we have, in the way that we are currently building artificial intelligence, you know, linear algebra and a shit ton of GPUs. Um, so, so I'm not, I'm not worried on that level about it yet.

[00:18:00] Um, I do think though some serious economic disruption for people in our industry is probably inevitable. The question is, does that disruption mean, does that disruption mean way fewer people? Are needed to do anything or will that disruption mean that, you know, you might need fewer straight up programmers, whatever we're going to call that, but you'll need way more people who have the experience and sort of defining and building systems and that the prevalence of AI will mean the cost for software will come down necessarily, which will allow the demand to go up.

I mean, and get software into places like it's never been, you know, your local lawn care guy might want a custom app. And before, that was ridiculously prohibitively expensive, but now, he can have it if it helps him and it's not gonna cost him a ton of money. So, that doesn't have to

Matt: be bad, but it is scary.

I think the only reason why I'm seeing success, I'll call it [00:19:00] success. Like, I've launched a few, like, stable apps using React and building something, like, super custom that I could never do without these tools. Just, not even, not even on the radar. So, I'll call myself successful, but the only reason why I've been successful is because I do have an understanding, um, of how the web works, how software works, and how I should de debug software.

So, I mean, it's still, like, right now, in the start of 2025, the only way I think people are going to have success with this stuff is if they have some Formed experience. And a total noob will not be able to use this stuff, especially if they're building something complex, right? Like, I'm writing tests and, um, utility functions, features for like these apps that I'm building.

I only know that because the developers I've talked to do that. Right? And I know that's how they're, that's, I know [00:20:00] that's how they're doing it. Right? So the, the average person is like, just gonna smash the keyboard, like, figure this stuff out for me. Yeah, maybe eventually it will grow, um, where it's doing it more native for you.

But to your point, um, of how this shifts the workforce, I still think you're going to have to have humans who are asking AI to solve a particular problem for them. Like, I don't think AI is going to get to the point where it's going to know exactly what you need. Out of software, I mean, I don't want to say never with this stuff, maybe a hundred years from now, but, I don't know, I just don't see it for my lifetime that we're gonna see AI just solve everything, and humans are still gonna say, ah, we're running into a problem over here for this lawn guy in Nebraska, it's, I know how to solve his problem, let me go build this software, AI's gonna do it for me, But because I'm the human with the experience and seeing what they're running into, I gotta [00:21:00] prompt it to tell it how to, like, what needs to happen.

So yeah, human and the human experience. Maybe that's like product. I'm just trying to put a label on it and just like thinking off the cuff. But maybe that's product managers, product, uh, product or project managers and product managers are going to be a better position in the future. Um. Yeah, just, you know, it is a crazy world to think about years from now.

Jonathan Williams: We don't have a name for what, what that job will be, right, because it combines what are now a lot of separate jobs. And so, you know, when we talk about disruption or, or, you know, there not being as many, um, not being as much room at the table, I mean, you know, that's a real concern. But it, but it's like you said, but I, I, and actually I, I would push back just a little bit on one thing.

I, I do think it's possible for noobs now, given where the tools have gotten in the last six months, I do think it's possible for somebody who doesn't know the lingo to build themselves [00:22:00] A useful piece of software. Well, you know, however we want to define that for the first time using only I provided, provided that that person has a lot of patience and, and actually here's another, another.

Bit of good news about AI. Provided that person has, A, a lot of patience, and, uh, B, is very literate, and just, just in terms of, like, articulating what they don't understand, you know, and also just being able to say, okay, well, you've used the term here, you know, you've told me that I need to, to whatever, install Node, I have no idea what Node is, or, you know, you've told me that I need to run a configuration script, how do I do that?

Um, but to be literate enough and patient enough to, you know, to go through and work with it, And, maybe, and the good news in that, is that literacy and patience have never been in great supply in the world, and they are only going down, as far as I can tell.

Matt: And I just want to jump in for an exact example of this.

I was working on Repl. [00:23:00] it, building something, actually no, I was using Bolt this time, and it was connected up to SuperBase because, This is going to get in the weeds, uh, but it was, it was using Superbase as the backend, which is a, I'll just call it a, a database service just from high level view. And it was asking me to, um, I was making some updates to this thing and it's like, okay, we're going to build this whole new RLS policy for you.

And I was like, What? Like, I don't even know what you're saying, and I spent the time asking, what is an RLS policy? How does this impact, you know, what I'm doing? If I make another feature request, are we gonna always have to make, you know, these RLS policy updates every single time I ask to do this?

Because I wanna know. How much of this I want to know how much of the unknown there is to me, right? Like I've already learned 20 percent up to this point and now you're going to introduce stuff. I have no idea. And I want to slow it down and understand what you're doing just so I [00:24:00] have a chance at survival when there's a, when there's an issue with this app that you've built.

Exactly. Right. So anyway, yeah, I just. Yeah, the literacy, prompting, having the conversation with the tool might feel weird, sound weird if you haven't done it yet, but it's super valuable.

Jonathan Williams: And then there's a, there's a resistance, or at least there was a resistance in me and still is a little bit. It's a, and I don't know if it's just a flavor of that, get these damn kids off my lawn kind of thing, you know, but it's like for those of us who.

Learn to code? Oh my god, people used to say that endlessly. Do you admit, like, for the last 20 years, any time, any time anybody brought up any sort of personal problem or financial problem, they were just like, learn to code. Well, some of us did. And now, and for some of us, it wasn't easy. Like, you know, I've met like alpha level coders and I'm not one of them and that's fine.

I don't really want to be, I mean, you know, being an alpha coder is great. If you want to create a brand new operating system or probably solve some of the stuff that they're doing at space X or advanced physics and stuff like that. But, [00:25:00] but even still, you know, and I. I still feel a little bit of that, God, was it like a waste to learn, you know, the, the syntax itself and all those, all those hours we spent pounding our heads, you know, getting sneered at on Stack Overflow, because unlike every other living programmer you've ever asked for help, AI has never seen a stupid question.

You know, it will, it'll, it'll walk you through it. And I also think that's great for people to remember. I'm ramming a little bit now, just that like, it's not going to judge you. It will not call you stupid. It will not tell you that, you know, you don't know anything you should give up, which is, you know, the most common thing you would hear on programming forums.

For a long time. So, oh, whatever our other worries, you know, might be, that's, that's incredibly exciting and encouraging, especially for, like, young people. I'm trying to get my kids into it, you know? And they will talk to that chat box endlessly because it will never stop answering.

Matt: Yeah, I know. Yeah, yeah.

That is one of the amazing things.[00:26:00]

I want to round off the conversation with a couple few, a few more questions here. So what's the future of, of WordPress for you? I'm a little worried that. Uh, we're not going to see the continued trajectory, the growth, the pie chart of WordPress grow as much as it has up until this point. I mean, I think we should feel pretty happy with the size of the market share that we have now.

But I can see a world where, listen, if things don't straighten out with WordPress and it's not better and sexier and people want to use it, I can see the market share go down, especially when somebody can just build whatever they want. In any coding language that AI spits at it. I'm not fully doomsday on that, but I do see a world where, and we could certainly trend in a direction where people go, Eh, what do I need WordPress for?

I got this over here and I'll just build it with, with this. What's your outlook on that? [00:27:00] So I'm

Jonathan Williams: more than a little worried long term for WordPress. And I want to be clear that, you know, me saying that I'm, I'm worried about something, or I think an outcome might be possible or likely, isn't the same as cheerleading for the outcome, right?

You know, there are plenty of things that I think might happen that I hope don't. Um, and this is in that case, you know, I'm very much a WordPress has been feeding my kids for quite a while now and I hope it continues to and I love the thing. That out of the way, WordPress is really poorly positioned at the moment for, to benefit from these AI tools.

And part of that is technical and part of that is community wise. I would, this is a, this statement isn't absolutely true, but I'm going to throw it out there because it's a. That's how I've been framing the issue. In many ways, right now it has never been easier to do hard things in WordPress. And it has never been harder to do easy things in WordPress.

Because you know, again, you can one shot a plugin right now with like a day's work [00:28:00] that you would have paid 2, 000 for months ago, right? Um, that people probably still pay 2, 000 for, if you know anybody who wants to spend that, you know. Give them my number. Um, but you can do it. Um, and that's amazing. And if, you know, big feature editions, and yet at the same time, if you want a simple brochure style marketing site.

Man, that's, that's a long road and I know a lot of people are going to jump in there and say no, you just need to use Bricks and no, you just need to use Beaver Builder and no, if you really understood how Gutenberg works and weren't complaining about it so much, you could do this in record time. It's like, well, kids, I've been doing this for decades and creating my current, uh, sales site, you know, oddjar.

com, my business site where I sell everything and advertise my services, creating that thing took longer than coding most of the plugins. Some [00:29:00] combinations, I think four or five of those plugins, testing, took longer than putting that site together. And you can say skill issue and that's fine. But the fact is Right now, if you want a simple brochure site, you can go to Bolt or Repl.

it or anything, throw in a library of pictures, throw in some copy, or just tell it what it does and ask it to generate all of that, and it'll do it. It'll use something like Next. js, which, yes, is overkill, and I think WordPress is better for lots of reasons, but it will spit that out in, like, you're done in a half an hour, man.

Um, and I know there's a lot of yes, buts behind that, but you know, the people who just want a website for their business, who historically, many of whom, if not most of whom would have gone to WordPress, there's an easier route for them now, and there's an easier route for the people building it, and I believe in our project as much as anybody, but we can't just keep going muh open source every time.

A [00:30:00] legitimate and in many ways better alternative emerges and, you know, it would also be nice if we could stop fighting so much on X and Twitter.

Matt: Yeah, I mean, I can't agree anymore. Thanks for coming to my TED talk. Yeah, no, there's a great, you summed it up, you know, I love that because I recently put out a video about, um, My experiences with AI on my own personal YouTube channel and like some of the comments, like one of the comments that stood out was somebody said like, Oh, AI is great because now I can replace those expensive form plugins every, you know, that I pay for every year with my own, with my own form that I make with AI.

My thing is, is once you start to use it, And if you're trying to let, uh, let's, let's just, you know, if you're trying to replicate gravity forms in one sitting, good luck, you know? And I, and I think what's going to happen, maybe, is that people will start using AI to code stuff. Let's say the beginner types, like [00:31:00] myself, or below.

And they're spent, like, you know, I built, my app took, it's, my app is just an RSS aggregator, that's all it is. It puts, takes RSS, it shows the words on a webpage, it's just built in React with a database backend, and I did some other stuff. But it took me like, 30 hours, you know, to make this thing. I, I think people are going to try to build their own software.

And spend many hours prompting and then realize, Oh, you know what? A hundred bucks at 150 bucks at 200 bucks for that plugin over there. I should have just bought that because it now has taken me so many hours to build this one thing. And this one thing can't be brought to the next WordPress website.

Because I didn't build it that way, because I don't know how, uh, and Gravity Forms is. So I think we might have, the fingers crossed moment for me is, there'll be a revelation in, oh, these price points, like your plugin, your add ons to Gravity Forms are anywhere between, what, 20 to 50, maybe 100 if you start buying from MultiLicense.

People are going to look at that and [00:32:00] go, I mean, why am I going to do this myself? Jonathan already did it and it's only 30 bucks. I'm just going to buy it,

Jonathan Williams: you know? And that's God, the, this conversation I have in my head every time I've at least something or, you know, do anything it's and I'm where you are.

Yeah, I know. I think that's, that, that's sort of where I am with. With, you know, buying software versus making it, you know, right now with the current state of AI. It's exactly that. It's that, you know, I, I probably could, if I wanted to spend several months with AI and not see my kids, create some sort of duplicate of Gravity Forms.

I mean, I could, I could make it do most of what I want it to do, right? But why would I do that? It's just, it would take so long to get anything, like you're saying, comparable and a lot of Add ons, and you could say this about WordPress itself, you know, with all the all the nastiness that's been going on back and forth between between wordpressers Online, you know, you know, just make your own I will I'll just work the damn thing and see you [00:33:00] know And it's just like

Matt: you could But you shouldn't yeah, you could but you shouldn't he's Jonathan.

His website is oddjar. com Oddjar. com it'll be in the show notes. The site does look great. By the way, I meant to say that Thank you very much From where you were the last time you were on the show. 'cause I know that was one of the struggles you had said was, yeah, it is like the marketing side of it, putting this website together, that's, yeah, that's a whole, you know, feat onto itself.

But your, your site looks great and I love, um, everything you've done with, with organizing that stuff. So check it out odd jar.com. Anywhere else you want folks to go to say thanks.

Jonathan Williams: No, that's it. Um, one other note is that, um, there's a, there's a free gravity forms add-on up there. It's, uh, list user entries.

And, uh, I think it's pay what you want, but you can put zero in that box and you will get that plug in. Um, and it's a simple but useful thing, I think. It just adds a list of each user's last 20 entries to any Gravity Form with links to their user, their admin user [00:34:00] profile. So that if you're in the user section and you want to bounce around between some links, it's a handy little thing.

Matt: We'll check that out. It'll be in the show notes, oddjar. com, list user entries for Gravity Forms. Check it out on oddjar. com. Thanks for hanging out today, Jonathan. Thanks for having me again, man. That's

it for today's episode. If you could do one more thing for me today, share this episode on social media, your favorite Facebook page. Group or Discord channel, spread the word about this podcast. It really helps. If you haven't added Breakdown to your favorite podcast app, point your browser to gravityforms.

com slash breakdown and click the icon of your app to add us and listen to us every two weeks. Okay, we'll see you in the next episode.