Write Laravel, not PHP (feat. Aaron Francis)

November 27, 2023

Watch on YouTube →

Transcript

Lane
00:00:00 – 00:00:01
What is PHP?
Aaron
00:00:01 – 00:00:11
PHP is the language that everyone makes fun of. So that's that's that's what PHP is. If you've ever heard somebody make fun of a language, the thing they're talking about is probably PHP.
Lane
00:00:35 – 00:01:04
Welcome back to Backend Banter, the show about back end technologies and careers. Today, I think we might have a little bit more banter than normal. I am famously not the biggest fan of PHP and Laravel, but one of the things I love to do on this show is bring on people that like other technologies and have different opinions, than I do so that you guys can get a lot of different perspectives. I don't want everyone just kind of regurgitating the stuff that I say, especially since, you know, a lot of it's wrong. So today, I'm joined by Aaron Francis.
Lane
00:01:05 – 00:01:07
Do you wanna take just a second and go ahead and introduce yourself?
Aaron
00:01:08 – 00:01:21
Sure. I am famously a big fan of PHP and Laravel, and I'm ready to banter about the back end. Yeah. My name is Aaron Francis. I work as a developer educator at a database company called PlanetScale.
Aaron
00:01:21 – 00:01:46
We are very good. So if you need a place to put your MySQL database, you should choose us. And I'm also a Laravel developer and have been for many, many years and love Laravel a lot. I make some YouTube videos, for Laravel, like, on in my spare time. And as, a developer educator, I make a lot of YouTube videos about MySQL, so do all kinds of stuff.
Lane
00:01:46 – 00:01:53
While we don't share a love for PHP, we do have quite a bit of common ground. I believe you're a are you a Vue fan?
Aaron
00:01:54 – 00:02:00
I am. I'm a big fan of Vue. That is definitely my front end framework of choice.
Lane
00:02:00 – 00:02:10
Yes. Me as well. Boot dev is written in Vue 3. Big fan there. Also, obviously, PlanetScale, we did an episode about PlanetScale 6 or 7.
Lane
00:02:10 – 00:02:35
Oh, that's probably been probably been about 10 episodes back now. And PlanetScale is used in our CICD course on Vudev, which is which is awesome. So, lot of common ground, but today, we're here to talk about PHP. And I'm at a huge disadvantage because I actually haven't done all that much PHP. I did a bunch of cramming for this episode, did a lot of PHP back in college, but it's been a while.
Lane
00:02:35 – 00:02:48
So you could probably even just lie to me about how great PHP is, and I'll just have to take it at face value. Can you tell us just let's just start with something really simple. What is PHP?
Aaron
00:02:49 – 00:03:02
Yeah. So PHP is the language that everyone makes fun of. So that's that's that's what PHP is. If you've ever heard somebody make fun of a language, the thing they're talking about is probably PHP. PHP is a super old language.
Aaron
00:03:02 – 00:03:44
Right? It's like one of one of the older languages that still exists. It's a I think originally, it was, like, truly a, you know, a templating language, and you would mix in your PHP with your HTML and just, like, throw up, you know, a script, and it would just load on the web server and kinda pull some stuff from somewhere and plop it into the HTML and throw it back to the client. So it was originally a scripting language, and it's just, you know, it's come a long way since then. It retains a lot of that simplicity, but we don't really use it as, like, a scripting language anymore.
Aaron
00:03:44 – 00:04:08
So, like, a lot of a lot of the criticisms I see about PHP are like, oh, look at this, you know, templating. I'm like, I just don't ever do that. Nobody ever does that. So one of the things I do love about PHP is that it's just, like, it's just productive, and it's just been around for a 1000 years, and I just, like, I love using it. So that's a little that's a little bit of background on on kinda what it
Lane
00:04:09 – 00:04:17
is. Well, I love the pitch of just productive. Like, I the the problem is I would argue that's why I use Go. Sure. But okay.
Lane
00:04:17 – 00:04:29
You you said you said something interesting, which was like Mhmm. Originally PHP was kind of scripting and templating, but now you use it in a different context. I'm guessing that context is like inside of, like, Laravel frameworks to build larger applications.
Aaron
00:04:30 – 00:04:46
Yeah. Yeah. Exactly. So I think now so, like, for example, in the Laravel framework, which is equivalent of Rails, but in the PHP world, in the Laravel framework, we in fact have our own templating engine, which is like, wait. PHP is a templating engine.
Aaron
00:04:46 – 00:04:55
And that's not like a lot of people are like, you fools. Come on. Come on, man. Like like, we know that off my list of Yeah. God damnation for later.
Aaron
00:04:58 – 00:05:13
And so we, like, we have this own templating language called blade. And so you'll see some of the criticisms online be like, oh, PHP is so insecure. Look. You could just dump this stuff out onto the page with this templating. And I'm like, yeah.
Aaron
00:05:13 – 00:05:24
But not nobody uses that. Like, nobody uses just regular PHP. We use blade as a templating engine. I think Symfony has maybe something called Twig. I'm not super familiar with Symfony.
Aaron
00:05:25 – 00:05:50
But, yeah, our entire, you know, our entire back ends are now all, you know, PHP much like you would have Rails. You know, we've got proper models, controllers, jobs, queues, all of that stuff. And then a lot of people don't even use blade as a templating engine. They throw it all over the wall to either an API or something like inertia JS, which kinda, like, binds the front end and back end together with a thin, like, protocol layer.
Lane
00:05:51 – 00:06:08
It's so interesting that you say, like, raw PHP raw PHP, templating isn't as popular. And the reason I say that is, I went to Python dot org today just to make sure I wasn't crazy. Yeah. And, I went to the, like, installation get started. And it was like, okay.
Lane
00:06:08 – 00:06:20
Download the Python interpreter, create a file dot main dot called main dot p y, you know, print hello world and then run Python main dot p y. And then I went to p h p dot net and it was like Not
Aaron
00:06:20 – 00:06:24
our best foot forward. I'm gonna be honest. Yeah. Here's the HTML
Lane
00:06:24 – 00:06:29
file, and it echoes a thing. And I was like, wait. How do I install PHP in the interpreter, like, superpowers?
Aaron
00:06:30 – 00:06:52
Many years it's been since I've written the word echo? Like, it has been a decade since I've written the word echo. And I think, like, I think, fundamentally, that's a big I think PHP.net is a big problem for us, frankly. PHP dotnet is ugly, and it sucks, and it looks like it's, you know, 15 years old, which it probably is. So that's a problem.
Aaron
00:06:52 – 00:07:22
But I think the other thing is, like, one of the weird things about Laravel is that it it it exists in the PHP world. And what I what I mean by that is, like, I wouldn't pick up PHP to just do a raw PHP anything, frankly. The the reason like, Python Python is a standalone language or Go as a standalone language. Fine. Great.
Aaron
00:07:22 – 00:07:44
Do all kinds of wacky stuff. But PHP is a standalone language. I wouldn't I don't I don't think I would use it. And so there's a real weird there's, like, a real weird dichotomy of, like, people trying to defend PHP as a language and comparing it to, like, you know, Python as a language. I'm like, yeah.
Aaron
00:07:44 – 00:07:54
Python's probably better. Like, I I I just you know, like, your exam your example there is true. Like, PHP dotnet, it's kinda lame, and it looks you know, it looks boring. Like
Lane
00:07:54 – 00:07:56
Laravel looks beautiful, by the
Aaron
00:07:56 – 00:08:44
way. Laravel on the other hand and that that I think is my like, that that's kind of the weird part is the Laravel, the framework has such a distance from PHP the language in terms of, like, like, productivity and developer experience and batteries includedness that I think the better, like if people are making a decision about what they wanna use, the better question is, like, do you wanna use Django or Laravel? And, you know, maybe people prefer writing the Python language and, like, that's fine. I don't really care about that. But, like, comparing the languages and then picking a framework seems like, I would I would I would view the ecosystems first before you decide that.
Lane
00:08:44 – 00:09:03
That's like that's almost a defining trait of frameworks in my mind. So so first of all, I wanna say, like, yeah, I get strong kind of like Ruby on Rails vibes in the sense that, like, very few people write Ruby without Rails. It seems. Yes. And it sounds like today few people are writing PHP without Laravel maybe?
Lane
00:09:03 – 00:09:06
Other than, like I guess there's kind of WordPress Yeah. Laravel.
Aaron
00:09:06 – 00:09:06
There's a
Lane
00:09:07 – 00:09:07
there yeah.
Aaron
00:09:07 – 00:09:22
There's an entire other community of which I am not a part and can't speak intelligently to, and that's the WordPress community. Don't know the first thing about it. Never written WordPress in my life. Prefer to avoid it if possible. And that one also gets that one also gets, like, thrown at us all out a lot.
Aaron
00:09:22 – 00:09:27
Like, WordPress code sucks. And I'm like, maybe. Beats me. I don't know. It's not it's not my community.
Aaron
00:09:27 – 00:09:40
Like, I I don't know. And to the Ruby thing, the Ruby as well always declare that Ruby is the most beautiful programming language in the world, and so it's like, okay, cool. Like, do you mostly use it with Rails, though? So Right. It it, like, seems
Lane
00:09:40 – 00:09:42
to only exist in that context.
Aaron
00:09:42 – 00:09:43
Yeah. Exactly.
Lane
00:09:44 – 00:10:12
So to your point about, like, choosing the framework first, this is actually really interesting because, like, I think this is a defining trait of frameworks. So, for example, in the Go community, there are very few, like, well, there's really no, like, rails equivalent or Laravel equivalent in the sense that it's like a a a fully featured framework that does all of the things, batteries included, right? Like authentication, database routing. Yeah, We tend to, like, compose libraries. Mhmm.
Lane
00:10:12 – 00:10:58
But obviously, Laravel, Rails, Django, these are kind of fully fledged frameworks. And I would argue that my definition of a framework is that like, instead of writing your application and injecting a dependency, you're like starting with the giant dependency and you're like injecting your code into it. And I would agree with you. Like, if you're using frameworks to build web applications, you probably should start with the framework because it's not something that you can, like, migrate away from. So, like, in the sense that, you know, if you're using some little some little authentication library, You can fairly easily swap that out, but it's not, like, trivial to swap out Laravel for some other PHP framework, I would imagine.
Aaron
00:10:59 – 00:11:10
No. No. I I I I wouldn't think so. And I think, yeah, I think that would be ill advised anyway. And I think the question about, like, what is a framework and what is not, I think your definition is correct.
Aaron
00:11:10 – 00:11:34
I think, however, we've gotten to this point where a lot of things, like I I think is specifically on the JavaScript side. I don't know the Go world at all, but I observe the JavaScript world mostly from the outside, and I see stuff like, Next being a framework. And I think according to the definition that you and I agree upon, Next is a framework. Yeah. Like, I I totally agree.
Aaron
00:11:34 – 00:12:04
Like, you write your code inside of next, and next handles, like, calling your code, not the other way around. And so I'm like, yeah, next is a framework. But then the term gets so muddied because people will start to say, like, well, what about next versus Laravel? And I'm like, that those are those are entire those are like orders of magnitude apart. And it'll take my take my, you know, fanboy out of it and say Next versus Ruby on Rails.
Aaron
00:12:04 – 00:12:16
And I'm like, those are fundamentally different things, and people are like, ah, I would much rather use Next than Rails, and I'm like, hang on. Okay. So Next is a framework. Sure. No, like, let's move on.
Aaron
00:12:16 – 00:12:25
No problem there. But then you start thinking about what do other frameworks have? What do Rails, Django, and Laravel have? Well, they have authentication. They have authorization.
Aaron
00:12:25 – 00:12:40
They have a database layer that's built in. They've got caching. They've got queues. They've got routing. They've got, like, all of this stuff, this rich ecosystem of everything that you need to build a modern web application.
Aaron
00:12:40 – 00:12:56
Like, how are you gonna send emails? Like, that's built in to Rails and Laravel. And I think something, like, something that people miss is they say, well, I'll just pick the best in breed package and install it into my JavaScript application. And you're like, yeah. Okay.
Aaron
00:12:56 – 00:13:08
So if we think about, like, a network graph. Right? So there are, like, 2, 3, 4 nodes in this network. So you've got 3 nodes, and they're all connecting to each other. You have a certain number of lines.
Aaron
00:13:08 – 00:13:11
Right? Certain number of edges between these nodes. You add a new node. The edges in explode. You add a new node.
Aaron
00:13:11 – 00:13:36
The edges just explode. So every node that you add, the complexity in the interaction between the nodes, this is like standard network stuff, like, increases multiplicity multiplicatively or exponentially. I don't really know which one. And that's how I view, like, that's how I view these build your own framework discussions. It's like, sure.
Aaron
00:13:37 – 00:14:05
You could decide. I'm gonna install, an ORM, and I'm gonna install an authentication authorization and an email sending and a background job and a scheduled task. Every one of those things you're adding is a new node in your in your graph of whatever your application is. And you you then become responsible for maintaining that connective tissue between every other node in the graph. And so then you're like, alright.
Aaron
00:14:05 – 00:14:21
Well, I'm installing this routing layer. How do I make sure it interacts with everything? Like, how do I make sure it interacts with, you know, route model binding if I wanna have slash team slash one? Like, oh, I gotta do that. And then every single node is on its own independent upgrade cycle.
Aaron
00:14:21 – 00:14:44
And so you're like, oh, well, my ORM upgraded to upgraded to 4.0.0. And now it's like, well, I gotta make sure that that's compatible with my routing and with my background jobs and with my auth. And it's like, y'all, this is like and that's where I think the framework discussion breaks down is like, sure. Sure. They're they're both frameworks, just like just like a bicycle and an airliner are
Lane
00:14:44 – 00:14:45
both modes of transportation. But I think
Aaron
00:14:45 – 00:14:45
we're just talking about fundamentally different things.
Lane
00:14:52 – 00:15:24
Doubting what you're talking about, just just go to your little Next JS project, and open up the node modules folder and immediately lose vision of every other file in your project. Like, this is how we end up with 70,000 dependencies in in node modules. Right? And I think there's 2, like, kind of solutions or approaches to this, like, exponential explosion of dependencies. The philosophy in the go community is like, well, let's just have fewer nodes.
Aaron
00:15:25 – 00:15:25
Sure.
Lane
00:15:25 – 00:15:48
And so there's like kind of the thought leadership in the go community is like, well, let's just use very few dependencies. We'll shove as much of we as much of what we need into the standard library as we can. Yep. And then we'll like add extra stuff. And And the Laravel approach, which I think I think is also a valid approach is like, well, we will just choose all of the dependencies for you and we'll manage them so you don't have to worry about it.
Lane
00:15:48 – 00:15:55
Right? The worst of both worlds would be like, okay, we're going to use all the dependencies and you're going to manage them.
Aaron
00:15:56 – 00:16:20
Yeah. And and the the insidious part is that spun is freedom. It's like, you get to decide, you get to build your own thing. And I'm like, I would rather I would rather build my business value than, like, cobble together a framework of, you know, 15 different packages that then are on 15 different upgrade cycles that I have to, like, manage. And so it's like, ah, I get it.
Aaron
00:16:20 – 00:16:25
Like, you're telling me that I'm free, but boy, do I feel, like, do I not feel free?
Lane
00:16:26 – 00:16:52
This brings me to, like, an observation that I've that I've seen is, like, every programming community on Twitter seems to be very concerned with engineering, which I think is a good thing. If you're trying to, like, build a career as an engineer, you know, you're talking about code. You're talking about writing better code, more performant code. You're weighing dependencies. The PHP community on Twitter just seems to be making money.
Lane
00:16:53 – 00:16:54
Like Doesn't seem that
Aaron
00:16:54 – 00:16:54
way, doesn't it?
Lane
00:16:55 – 00:17:03
To be all the people that are actually making money, even as, like, indie businesses. What is that about?
Aaron
00:17:04 – 00:17:27
Yeah. It it does seem that way, doesn't it? I think so I have I have several opinions about this. One, I think, is that all of the money, like, all of the institutional money has flowed into JavaScript. And so I think there's been a lot of, like, time on people's hands to do stuff that maybe moves the needle by a microsecond or 2.
Aaron
00:17:27 – 00:17:55
And it's like, hey. You've got, you know, 7,000 engineers at, you know, Facebook working on React, and that's why we have the mess that we're in. And it's like, ah, you know, PHP doesn't really have that luxury. We've never been we've never been, like, the cool kids, and we continue to not be the cool kids. And so it's like it's almost like it's like PHP slash Laravel is like a 1994 Ford F150.
Aaron
00:17:56 – 00:18:23
And you're like, that mother freaking truck will not die, and I just drive it from the farm to the market every day for the past, you know, how 25 years, 30 years, whatever, and it just works. Meanwhile, like, the I feel like the the JavaScript community is very, very, very invested in tinkering on their vehicle instead of, like, where are you like, where should the vehicle go to sell the thing that you're trying to sell? And they're like, no. No. No.
Aaron
00:18:23 – 00:18:54
Like, if we make the steering wheel if we make the steering wheel go up by an inch, like, it's a little bit more ergonomic, and you're like, you gotta drive the car, man. And so I I do kind of feel I do kind of feel like because we're not the cool kids, which I admit we're not, and we don't have all the VC money pouring into, like, our space, We don't have the, like, we don't have the luxury of side questing very much. And so we we just don't end up on a lot of side quests. We're like, hey. This is really good.
Aaron
00:18:54 – 00:19:13
I'm gonna use it and then ship whatever whatever the end goal is. So I think people are focused on different goals, and a lot of the JavaScript community is focused on the tooling and, like, the internal, almost like navel gazing, and the PHP community is more like, I gotta put food on my table. And so we're like, I'm just gonna look down the road.
Lane
00:19:14 – 00:19:26
Yeah. That's that's that's definitely how it looks from the outside to me. And as as someone who does like tinkering much to the dismay of, like, my wallet, essentially. Mhmm. Yeah.
Lane
00:19:26 – 00:19:34
I I I look at PHP as a language, and I go, ugh. But then, like, the other wheel drives a Lambo. So, you know Yeah. It's fine.
Aaron
00:19:34 – 00:19:45
Yeah. And it's not to say like, it's not to say that one or the other is better. Right? Because it's it's a it's a silly discussion to have outside of context. Right?
Aaron
00:19:45 – 00:20:07
So those 7,000 engineers inside of Facebook, like, their focus and their job, 1, is to get paid by Facebook and to improve React or whatever. I made up 7,000. I have no idea how many people work on it. They're that's that's their end goal. And I think what happens is I think what happens is the message gets delivered without the context of the end goal.
Aaron
00:20:08 – 00:20:36
Right? And so on Twitter, we see a lot of thought leaders who are building like, who are actually building the tooling, and their end goal is to build and worry about the tooling. But we don't have that like, we don't make that context front and center enough. And so what happens is people who are then trying to build and deliver products look at all of these thought leaders, and they're like, oh, they're building and focusing on tooling. I should focus and worry about tooling myself.
Aaron
00:20:37 – 00:20:38
And it's like, no. No. No. No. Wait.
Aaron
00:20:38 – 00:21:08
Y'all have entirely separate goals. Neither goal is inherently bad. But when you adopt the methodology of someone who's trying to go for a different goal than you are, you're gonna end up spinning your wheels. Like, you're gonna end up wasting a bunch of time because you're operating by a set of principles that will not get you to the place you're trying to go, and it will get to the place they're trying to go, which is to build better tools. And so it feels like it feels like the the discussion on Twitter is not nuanced enough, which shocker.
Lane
00:21:10 – 00:21:20
Yeah. Who would have thought? Hey. The the community is like, I'm starting a business as, like, an entrepreneur. I'm getting into coding as a front end developer.
Lane
00:21:20 – 00:21:39
I'm learning DevOps. I'm doing machine learning. Like, all those communities kind of have huge overlap on Twitter. And, like, everyone's goals are different and everyone's arguing about how to get to their own goal. And you could be an indie hacker, right, trying to build, like, a side project and arguing with some of this, like, trying to get a job at Facebook.
Lane
00:21:39 – 00:21:46
And, like, your goals are different, and you're arguing about technology. And it's yeah. I I completely agree. That's, it's actually a big problem.
Aaron
00:21:47 – 00:21:59
Yes. Huge problem. It's so silly. It's so silly. Like, why would I why would I argue with someone about how to build a house if I'm trying to whittle, you know, a statue or something.
Aaron
00:21:59 – 00:22:14
It's like Right. We're talking about different things. We're both woodworkers. You're building a house, and I'm whittling a little knickknack for my child. Like, we don't really need to be having a discussion, and yet that context is lost and we're arguing about which tools to use.
Aaron
00:22:14 – 00:22:17
And I'm like, there's different use cases, man. Yeah.
Lane
00:22:18 – 00:22:31
So I'm just going to go ahead and yoink the use case right over to, engineering. We're gonna forget about indie hacking for a minute. Great. And I'm gonna bring have you ever read the article PHP
Aaron
00:22:31 – 00:22:32
Oh, yeah.
Lane
00:22:32 – 00:22:33
A fractal of bad design?
Aaron
00:22:34 – 00:22:36
Yep. Super, super duper have.
Lane
00:22:36 – 00:22:41
Yeah. So so this was my introduction. I'm not gonna make you, like, counterpoint the article. I didn't No. No.
Lane
00:22:41 – 00:22:47
No. Article. I'm just I just want to reference the article because, like, this is one of my first introductions to PHP in college.
Aaron
00:22:47 – 00:22:48
Sure.
Lane
00:22:48 – 00:22:57
Was like, okay, we're going to do some PHP includes. And also, my professor shares this article and says, oh, also PHP is an awful language and you should probably try to not use it.
Aaron
00:22:57 – 00:22:57
And
Lane
00:22:57 – 00:23:14
so I go, ah, yes, PHP is bad, right? Like, I'm just told that's why my professor, and I just kinda eat it, hook, line, and sinker. Mhmm. Fast forward 8 or 9 years, and I'm, like, hearing about PHP again. And anytime PHP is mentioned online now, all I hear is we've changed.
Lane
00:23:15 – 00:23:22
Like, it's Yeah. Everything has changed. No one writes PHP like that anymore. What are what are some of the big changes that have happened over the years?
Aaron
00:23:22 – 00:23:36
Yeah. So PHP is fractal of bad design. A scathing article about all of, like, the warts and stuff of of PHP. And, you know, to some extent, like, meh. Yeah.
Aaron
00:23:36 – 00:23:47
Like, I I get it. Sure. Yeah. We we got we got some weird stuff. There are some bad there are some bad language decisions in PHP, and I'm not gonna, like, try to tell you that they're good.
Aaron
00:23:47 – 00:24:14
Like, some of the some of the funniest ones are, like, in some in some, function calls, it's haystack needle, and in some function calls, it's needle haystack. And it's like they they swapped the arguments around for, like, array searching and other stuff. And so people will look at that and be like, what a disaster. What a what a unserious language. Who's driving here?
Aaron
00:24:14 – 00:24:21
Yeah. And yeah. And I'm like, sure. That's not ideal. I agree with you on that.
Aaron
00:24:22 – 00:24:51
However, that like, the haystack needle needle haystack mismatch was probably written in the late nineties or early nineties. And I think to their great credit, like, there's some there are some very good people driving the language itself. To their great credit, PHP is extremely backwards compatible. And you can imagine the kinds of the kinds of Python 2, Python 3 Fractal we would have when we're starting to talk about PHP. Right?
Aaron
00:24:51 – 00:25:15
Because PHP is everywhere. It's like, what, 70% of the web and, you know, maybe 50% of that is WordPress or maybe even more of that is WordPress. And you can imagine if we start mucking around with, like, oh, actually, that one function is now the opposite. Like, we switched the order so that we would make some people online happy. It's like, I don't know what the value of that is.
Lane
00:25:15 – 00:25:17
The entire Internet breaks is the consequence.
Aaron
00:25:19 – 00:25:26
Internet starts returning, like, negative one because it's, like, I didn't find that I didn't find that haystack in this needle.
Lane
00:25:26 – 00:25:27
You're like, no. That's not
Aaron
00:25:27 – 00:25:54
what I meant. And so I think I think beyond that, though, is, like, that article is a tear down and a critique of raw PHP, and I would again say, like, I just don't use it. I just I just don't use it. Like, so and even from even from then, PHP's package manager is far and away one of the best ones in any ecosystem. Package manager is incredible.
Aaron
00:25:54 – 00:25:56
We only have one of them, which is
Lane
00:25:56 – 00:25:57
great. Composer?
Aaron
00:25:57 – 00:26:04
Like, everyone uses composer. That's right. Everyone uses it. It handles file auto loading. It handles namespaces.
Aaron
00:26:04 – 00:26:20
It handles everything really nicely for us, like dependency resolution. You can't have like, I think in JavaScript, you can install, like, 6 versions of the same dependency if different packages require it. Yeah. And composer just says no. Like, no, you've got to figure out which one you want because this is nonsense.
Aaron
00:26:21 – 00:26:34
I'm not going to allow that. And so, like, the ecosystem itself has moved forward a little bit. Like, nobody's do nobody's writing include this file dot PHP anymore. We don't do that. We just use composer auto loading.
Aaron
00:26:35 – 00:27:11
And one of the great things I think about so if we stay at the PHP level and don't talk about Laravel, I think one of the great things about the PHP level is there is an organization that exists to create standards for PHP that allow frameworks to interrupt nicely. And so we have this framework interoperability group. And the whole point of the PHP fig is what they're called framework interoperability group. It's like this this group of people, basically group of big PHP people. You know, Taylor Otwell is one of them, and then there's, like, you know, 50 or 60 more.
Aaron
00:27:11 – 00:27:49
And the whole point of that group is to be like, hang on y'all. Like, we have got to decide what what the interface is for an HTTP request. So what is the what is, like, the interface that we're all going to agree upon and adhere to, and what is the interface for an HTTP response, and what is, what is our, like, file auto loading format going to be? Because what we don't wanna do is we don't want Laravel to have, like, this way that it autoloads files and Symfony to have a way that it auto loads files. Right?
Aaron
00:27:49 – 00:28:01
We don't want that. That's not good for PHP. That's not good for Laravel. That's not good for Symfony. And so we have this group that proposes these standards and, like, code style is one of them.
Aaron
00:28:01 – 00:28:20
Like, we have a a PSR that just says, like, here's how you should style your code. And now we can all load that across the entire ecosystem, like, hey. This code is PSR 2 compliant, and, like, we just use PSR 2 styling. That's just what we do. And we can all say, like, hey.
Aaron
00:28:20 – 00:28:50
I wrote I wrote a caching library. It adheres to PSR, whatever it is, PSR 8. And so as long as you as long as your framework, like, will take a PSR 8 interface, my caching library is gonna work. And so what that allows us to do is truly, I think, push forward really, really good packages and libraries because we're not like, oh, I need to write, you know, I need to write a caching library for Laravel and a a caching library for Symphony or an HTTP layer for whatever. It's like, no.
Aaron
00:28:50 – 00:29:03
It's just, you know, it's just PSR 8. And so I think that's one of the great things. I would like actually to see that in the JavaScript world for them to, like, sit down and decide, hey. Let's talk about module resolution. Let's talk about bundling.
Aaron
00:29:03 – 00:29:20
Like, let's make some decisions that we're all like, all the framework authors get together. Let's make some decisions and decide, like, this is nonsense. We need to we need to be on the same page here. And so as far as a fractal of bad design, yeah, we've got a lot of warts. I will admit we have a ton of warts.
Aaron
00:29:21 – 00:29:52
We've got great backwards compatibility and we've got a whole organization that's focused on like improving the language and a different organization that's focused on improving the ecosystem. Not to mention you move up to something like Laravel and it papers over all of those problems because it's got string helpers and array helpers and it's got all of those things. And it's like, yeah, we'll just you know, we'll hide the warts because we can because we're a level up from the language that must maintain this, like, 30 year backward compatibility.
Lane
00:29:53 – 00:30:15
Yeah. I for any JavaScript listeners in the audience, if you heard PHP a fractal of bad design and you're chuckling to yourself, JavaScript's just as bad. Like, it's not you can't if you're a JavaScript fan, you can't be dunking on PHP. Yeah. And Python does famously have, like, the worst backwards compatibility ever.
Lane
00:30:15 – 00:30:15
The 2, 3,
Aaron
00:30:15 – 00:30:18
you all have that was big schism. The 2, 3 schism was bad.
Lane
00:30:18 – 00:30:35
It was really bad. And like, we still see like when I started learning Python back in 2010, I was like just getting like I was on Python 2 for the most part and then there was Python 3 and like in 2023, I'm still dealing with like Python 2 crap.
Aaron
00:30:35 – 00:30:36
Not not ideal.
Lane
00:30:36 – 00:30:47
It's pretty crazy. Okay. So we we can put PHP, the language itself, behind us for just a second. I know you'd obviously rather talk about Laravel. It's the thing you do.
Lane
00:30:47 – 00:30:54
Yeah. Big time. Tell me, like, let's just start with what's your favorite thing about Laravel?
Aaron
00:30:55 – 00:31:34
Yeah. So I think my favorite thing about Laravel is a little bit of a meta principle, and it is the fact that there is a Laravel LLC company that pays and employs, like, I don't know, 8 people now or something. And the entire purpose of Laravel LLC is to make Laravel the best ecosystem in the world. And so it is it is crazy to me that so, like, imagine if, you know, you've got rails and you've got DHH. DHH is really focused on base camp and making a $100,000,000 a year or whatever it is.
Aaron
00:31:34 – 00:31:51
Right? Which is awesome. Imagine that, but instead of DHA have DHH doing base camp, DHH solely focused on everything that could ever be wrong with the rails ecosystem. Boom. That's what Laravel is.
Aaron
00:31:51 – 00:32:13
Right? So what that like, it's a meta principle because how that trickles down is Laravel has answers for everything. And you may, like, you may decide, like, I actually don't wanna host on the Laravel the official Laravel hosting platforms, which are, you know, they're paid platforms. That's fine. There's still there's still other options.
Aaron
00:32:14 – 00:32:55
And I would say maybe 5 or 10% of what Laravel LLC offers to the world is paid, and the rest is just fully open source. And so, like, you start all the way back at local development. We have first party maintained, like, local development answers for getting Laravel, getting PHP, MySQL or Postgres, Redis, DNS mask, all of this stuff set up on your local machine such that you can buy you can walk out of this part is Mac only. You can walk out of the Apple Store, open your machine, download Laravel Herd, and then literally 3 minutes later be developing a Laravel application. No homebrew, no Docker, nothing.
Aaron
00:32:56 – 00:33:15
And then it's like, oh, you like Docker? Well, we've got an official Docker solution. Oh, you're on Windows? We've got solutions for that. And so it just kind of builds from there that it's like every single thing that you might need, there's there is an answer, and they're continuing to push the bounds of, like, what's available for free, which is is kind of nice to me.
Aaron
00:33:15 – 00:33:18
So that's like the meta answer about the ecosystem.
Lane
00:33:20 – 00:33:38
I loved this. I I jumped into Laravel again in preparation for this episode and downloaded, like, a few of the starter projects. And they're, like, maintained by the core Laravel team. So you're not just, like, kind of searching GitHub, trying to find random projects Right. Random it's like this is the recommended way to do it, which I think is fantastic.
Lane
00:33:39 – 00:34:01
Go is not the same in that it's not like a company behind Go specifically. It's like Google sponsored. But a lot of the ideas of, like, we have this benevolent dictator, benevolent organization that's doing a ton for the community building, like, official tooling. I think that's fantastic for the ecosystem. And I'm going to quote, I think our mutual friend, Dax, Dax Rad.
Lane
00:34:01 – 00:34:30
He was a guest on the pod a while ago. I'm actually not gonna quote him because I'll fuck it up, but I'm going to say things that he kind of has said in the past, which is I'm a big fan of these companies that like, for profit companies that are building in the open source space and have some sort of monetization strategy. Right? So with Laravel, obviously, like, Laravel's free and open source. There's always, like, additional tooling and help and hosting that you can get.
Lane
00:34:30 – 00:34:47
Yeah. It's paid because it gives me assurance that, like, when I invest my company's time and resources into this stuff Yep. That it's going to be maintained. Because you can have, like, a free open source package that you use, and and I use ton of these as well. But, like, it just stops being maintained.
Lane
00:34:47 – 00:35:01
The maintainer is like, I don't have time for this. Like, this is the thing I've just been doing for free. So I am I am definitely not against people building good businesses around something open source. In fact, I prefer it that way sometimes.
Aaron
00:35:03 – 00:35:26
Yeah. And I think what you see you see kind of that on the other side from DHH is, like, they're continuing to push Rails forward as they extract stuff out of Basecamp. So they kinda have a similar model but different. And then they have, like, a full on, like, steering committee and, you know, core team and all of that. So they're they're kinda mirroring the same structure, but there's there's gotta be there's gotta be an incentive.
Aaron
00:35:26 – 00:35:33
Right? There's gotta be money. Like, it's grass maybe, but there has to be money or it stops. Like, what do you expect?
Lane
00:35:33 – 00:35:52
Yeah. I it can't be any other way. Like, people will only do so much for free for so long. Like, we all eventually need to be paid to work on stuff full time. Like, unless you're independently wealthy and just love doing it, pretty crazy to have other expectations.
Aaron
00:35:53 – 00:35:55
Yeah. Yeah. I think so.
Lane
00:35:55 – 00:36:03
Okay. So we got what you love, which is, like, the ecosystem and the company and the organization behind Laravel. What's a thing you don't like?
Aaron
00:36:05 – 00:36:06
Man.
Lane
00:36:09 – 00:36:13
If you say nothing, all credibility goes out the window. I know. That's why
Aaron
00:36:13 – 00:36:47
I really don't wanna say nothing. I I I'll be honest. I have a really hard time I have a really hard time saying something I don't like about Laravel. And again, I I think I approach it from a very almost, like, lowbrow workman point of view, that, like, this is the best hammer I've ever seen in my life. It's like, yeah, but it's it's a hammer.
Aaron
00:36:47 – 00:37:05
Like, don't you wanna, like, have I'm like, no, man. I can nail so many nails with this thing. And so I have a hard time even thinking about, like, what's what's terrible about it. I think some of the common critiques are, it's too big, it's bloated, it's slow. And, like, sure.
Aaron
00:37:05 – 00:37:21
Yeah. Like, there's a huge there's a huge surface area. And I will say I have not been a Laravel beginner in a very, very long time. And so maybe it is true that, like, the fact that we have all the batteries built in means you have to learn a lot of batteries. And maybe that's like maybe that's overwhelming.
Aaron
00:37:21 – 00:37:22
I can't really speak
Lane
00:37:23 – 00:37:26
to that. There was a lot of files when I spun up the getting started. Yeah.
Aaron
00:37:26 – 00:37:28
Yeah. Yeah. For sure.
Lane
00:37:28 – 00:37:28
Yeah.
Aaron
00:37:29 – 00:37:51
Yes. There are there can be a lot of files. There can be a lot of, concepts. Like, you've got, you know, built in, you've got queues and scheduling and mails and, you know, routes and middlewares and all this stuff built in. So I can see that being like, woah, dude, I just want to freaking start an app here.
Aaron
00:37:51 – 00:38:15
So I can see that being a problem. And then the speed, I understand that a lot of people will look at that and be like, oh, PHP is so slow, isn't it? Especially compared to something like Go, for example. And my point of view on that is just so like, man, that's a champagne problem. Like, I personally am just not like, I just don't have that problem where I'm like, I've got to serve 50,000 requests in a second.
Aaron
00:38:15 – 00:38:18
Otherwise, I'm gonna lose $1,000,000. It's like But to
Lane
00:38:18 – 00:38:18
be fair, it's like I
Aaron
00:38:18 – 00:38:20
would Yeah. I would love to have that problem.
Lane
00:38:20 – 00:38:24
You're making money hand over fist when you start having Yeah. That would
Aaron
00:38:24 – 00:38:53
be awesome. So and I think, you know, I think that a lot of that is countered by, frankly, just, like, throwing more money at it. PHP has this really great request response model where it just dies. Like, the entire universe is constructed upon request and the entire universe is destroyed upon response. And so it's really, really, really easy to fit that into like a serverless architecture, for example, because that serverless is whole deal is like one thing at a time.
Aaron
00:38:53 – 00:39:09
And PHP is like, Yeah, I've been doing that since the nineties. And so there are ways like there are there are ways to get around, like some of the slowness of PHP. But frankly, I'm just not like I'm not a successful enough SaaS creator to have ever worried about PHP being too slow for me.
Lane
00:39:11 – 00:39:26
I love that. Yeah. The the context I've worked in in my career, I've certainly never created a company that's needed the scale the the speed that Go offers. I've only created one company, and we don't have that scale yet. But I have worked at companies that have needed it.
Lane
00:39:27 – 00:39:43
Yeah. And it's it's typically not in the sense at least in my experience, the companies I've worked for, it's not been like we have so many users. We can't serve a web page fast enough. That's never the problem. It's like we're processing the entire Reddit pipeline of data.
Lane
00:39:43 – 00:39:46
Like, that's that's the problem that we're solving with. Yeah. Typically.
Aaron
00:39:47 – 00:40:15
And I've I've run into problems with like that with PHP, where it's like I get a 4 gigabyte CSV file from a vendor and I got to churn through it. And then it's like even even then the database has become the bottleneck for me. It's like, I can I can just spin up 8 pro 8 PHP processes, which is what I did? It's just, you know, fork out a whole bunch of processes and run the same thing. And if I were more sophisticated, I would have maybe, you know, written that particular part and go.
Aaron
00:40:15 – 00:40:26
But again, I'm just I'm just a workman. Driving F-one hundred and fifty is like, oh, it works. I'm just, yeah, we'll keep moving. And so, yeah, that is a valid critique of PHP and maybe Laravel specifically.
Lane
00:40:27 – 00:40:36
Yeah. Well, like if your process is parallelizable, right, then you can just kind of throw money at it. Yep. And it tends to not be as much of a problem. So I totally get that.
Lane
00:40:36 – 00:40:44
Okay. So I do have some specific questions about things you may or may not like. Yeah. Yeah. Although, now I'm realizing they might just be PHP questions.
Aaron
00:40:45 – 00:40:48
The first is I can talk I can talk to both sides, so just hit me with
Lane
00:40:48 – 00:40:54
what you got. The first is dynamic typing. Is the are you a dynamic typing fan, or does it bother you?
Aaron
00:40:54 – 00:40:55
Live for it.
Lane
00:40:55 – 00:40:58
Love it. So Just give me
Aaron
00:40:58 – 00:41:23
give me the freedom of turning an integer into a string and back and forth again, and I'm a happy man. So here's the deal. One is I have to expose my biases. I don't like TypeScript. I feel like it's it's like a I'd say it is a hoop jumping practice to satisfy something that is not really a problem.
Aaron
00:41:23 – 00:41:31
So now that you know that, listener, you can disregard everything else I say because I'm not a trusted person. Typescript here. You're okay. No. K.
Aaron
00:41:31 – 00:41:55
Good. So typing in PHP, can be very gradual, and there are certain there are instances where I very, very much like it, and there are instances where I super duper don't care about it. And so, like, let's say, like, working working in a, you know, web application, you've got a user model most of the time. Right? You've got a blog post model most of the time.
Aaron
00:41:56 – 00:42:21
You know, in PHP, those are proper like classes. So that's the first thing is we're not having these we're not we're not usually passing around objects that are shaped a certain way but don't have any enforcement. Right. So I feel like TypeScript maybe exists because in the JavaScript world, you do a lot of like JSON object passing and it's like, what's the shape of the object? Nobody knows.
Lane
00:42:21 – 00:42:21
Right.
Aaron
00:42:21 – 00:42:32
And then in PHP, it's like, hey, give me a user model. And you know, like, you know what the user model is. And so that's one that's one place that I like to do. You know what the What's
Lane
00:42:32 – 00:42:39
the difference? Like, you say you know what the user model is. Do you get, like, what I say user dot, do I get, like, autocomplete?
Aaron
00:42:40 – 00:42:52
How does that work? You do. You do. And, well, you wouldn't in PHP because you'd have to use the arrows, not the dots, but, that's a little bit, that's a little bit of a quirk that people hate also. So, yeah, you do.
Aaron
00:42:52 – 00:43:08
So, like, let's say for example, I have, like, a login function, and the login function has a argument or a parameter called user. Right? So just log in parenthesis user dollar sign user. But that's PHP. Right?
Aaron
00:43:08 – 00:43:22
So log in dollar sign user. And at that point, user can be anything. User can be a number. User can be a string called foobar. I don't love that if I know, as the developer, they need to give me a user object.
Aaron
00:43:22 – 00:44:20
And when I say user object, I mean a class instance of a user. So what I will do is I will type hint that- I'll type hint that argument and say, this argument must be an instance of the class 'user.' And so then PHP will enforce that whatever argument is incoming to that method is an instance of an actual class that you can, like, command click into and go to the class. And it's like, here's the user class definition. And so instead of just being, like, a pretend object shape that we're all agreeing on to call user, it's like, where's the user dot PHP file where I can go and look at the properties, the methods, all of the affordances that are given the user object? And so I feel like that that is a that is a big difference, is I'm not maintaining, like, a type definition of, user has a first name and it must be of string.
Aaron
00:44:20 – 00:44:56
I am I, like, have a user class, and the user class exists out of necessity, and I'm not creating some extra layer on top. It like, what how else am I supposed to write a program without a user class? And so then I can enforce that when you pass me a post to be published, that that post is of instance type blog post. And then once I'm inside that method, I can call, you know, arrow publish or arrow delete because my, you know, my IDE will auto complete, like, hey, I know that this is a class. I'm gonna go look at the class and see what methods are there, see what properties are there.
Aaron
00:44:56 – 00:45:29
And so that's one area where I really like, you know, gradual typing to say, if you're gonna if you're gonna tell me to log somebody in, it better be a somebody and not a a string called foobar. And then actually inside of those classes, like, if you have a user class, for example, you can have, like, class level properties, obviously, of, like, let's say, you know, name, email, whatever. And you can declare those in in the class. You can declare those as strings. And so then if you try to set the user's name to the, you know, an integer one, it's just not gonna work.
Aaron
00:45:29 – 00:45:55
Your IDE is gonna tell you, but it's also just not going to work. And so you can gradually type these things, but there's no, like, there's no maintaining of some external definition that can get out of sync. It's it's, like, all kind of inferred out of the work that you're actually doing. And now do I, like, type all of my variables and say, like, dollar sign name equals quote Aaron and, like, somehow enforce that that's a string? No.
Aaron
00:45:55 – 00:46:04
No. I don't do that. But, like, at the class level or the object level, yeah, I put some types on it and I'll pass them around to know what my affordances are, what my methods are, that kind of thing.
Lane
00:46:05 – 00:46:11
Here's my hot take. I think you might like static typing. You just don't know it. Here's the
Aaron
00:46:12 – 00:46:15
the I I think that that could be real. That could be real.
Lane
00:46:16 – 00:46:38
Let let me explain. So it sounds like and again, I'm not a PHP expert, but sounds like PHP has a system kinda like Python, where, like, you can add type hints, and they're not really enforced by, like, the interpreter, but your IDE picks up on them and will kind of give you the autocomplete and the, hey, you're using a string here. Is that does that kind of sound like what the language that
Aaron
00:46:39 – 00:46:44
the PHP will refuse to run. It'll throw an exception if you pass the wrong time. So there is there is safety. Yeah.
Lane
00:46:44 – 00:46:52
So it is strong. Okay. Stronger than what Python is doing then, I think, because in Python, it like, the interpreter ignores it, but essentially your ID will pick up on that.
Aaron
00:46:52 – 00:47:03
Yeah. I think as long as you've declared strict types, which is a PHP option, which I think all frameworks declare strict types, it won't coerce the types for you. It'll just it'll error out.
Lane
00:47:03 – 00:47:31
Got it. Okay. Yeah. So this sounds to me like at least some level, some like if you're like if we can treat static typing as not like a binary thing, but more like a gradient, then it's definitely some level of static types. And it sounds like the thing you don't like about type of static typing is actually just a TypeScript specific thing, maybe, because I hate it too, which is like you have this like types, and you have to maintain your
Aaron
00:47:32 – 00:47:56
types. Yeah. And I don't like, like, I don't like the gymnastics, I think, of TypeScript where I I feel like there's a completionism streak in in a lot of TypeScript people that's like everything must be statically typed. And if I can get everything typed just right, I won't have any errors. And that's just not been like, that's just never been the case for me.
Aaron
00:47:56 – 00:48:33
And, like, the frankly, types have never been my biggest source of bugs. Like, logic bugs have been my biggest source of bugs where I'm like, oh, I was actually off by an hour in that calculation. Like, that is just I freaking that wouldn't help me with had to have TypeScript. And I feel like one of the problems with, TypeScript from my point of view is all of these amorphous blobs of objects. I understand that, like, you need visibility into what blobs are being passed around, and we just don't have huge amorphous arrays that we pass around in PHP.
Aaron
00:48:33 – 00:48:58
And, frankly, if if we did, I would probably put that into it, like, a class on its own. I'd make a data transfer object. I'd make a DTO that is, like, more structured, frankly, just so I can get the IDE auto completion. Like, I just I just feel like there's a difference in the way the language works on the PHP side where we're very, like, object oriented, And PHP side is like, here's a JSON object. Like, good luck.
Aaron
00:48:59 – 00:49:03
So I'm with you again, and I but I, let me
Lane
00:49:03 – 00:49:28
let me explain. So, when you say I would use a class in my mind, what I hear is I'm I'm using static types in the sense that, what is a class, if not essentially a definition of like properties and methods that exist on a thing? And I feel like it's the comparison to TypeScript that's hurting. God damn TypeScript. Like, because in TypeScript, you just like make an object literal.
Lane
00:49:28 – 00:49:38
Like, you don't, like, create a class for it. You just, like, open curly brackets and start typing stuff. And then in TypeScript in its own annoying way, you have to, like, go make an interface for it. Right?
Aaron
00:49:38 – 00:49:38
Right.
Lane
00:49:38 – 00:49:58
Whereas in Go, what I would do is create a struct, which is essentially for the purposes of structuring data, works like an object. Right? I create the struct definition and then I make a bunch of instances of the struct. It sounds like in PHP, you do essentially the same thing. You're creating a class and then you'll create instances of the class and you get auto completion on it.
Lane
00:49:58 – 00:50:27
And in my experience, when I don't have that, just the struct or the class or whatever with with type hints, I do get a ton of type errors, in the sense that, like, I misspelled the property name or something and my ID didn't yell at me. But if you take that away, like, I don't use 90% of TypeScript's features on the front end. I really just, like, want to know what's in my stinking object. Like, I want to know what properties.
Aaron
00:50:27 – 00:50:43
Yeah. Yeah. And I don't know. Like, I don't I don't have any I don't have any grandiose, like, opinions or points of view on, like, static typing in general. Like, I'm just not a very sophisticated developer, frankly.
Aaron
00:50:44 – 00:51:02
What I do have what I do have points of view on is typescript, and it looks like it sucks. Like, it just looks hideous. It looks like a pain, and it looks like it sucks. And I just don't I just don't wanna use it. And if you can trick me into doing typing like PHP has, like, yeah, I'm a simple man.
Aaron
00:51:02 – 00:51:30
Like, if I'm if I expect a user, I'm going to tell PHP this should be a user. And, like, I don't go super far into creating DTOs so that everything would be would be typed. But if I start manipulating, like, if I start manipulating a multidimensional array and in our in in PHP, an array can be like a hash map in Ruby. So you can have, like, named keys, which is basically just like it. It does both.
Aaron
00:51:30 – 00:51:52
Yeah. So it supports numeric indexes just like an array, and it supports named keys like a JSON object or a hash map. Right? So if I start creating this really big associative array, which would be like a hash map, and I'm starting to shove all these different settings and keys in there, I would frankly, I'd be like, oh, this is weird. And I wouldn't think, like, I'm abusing like, I'm betraying the gods of types.
Aaron
00:51:52 – 00:52:11
I'd think, like, oh, this is kinda nasty, isn't it? And I would create, like, a user settings DTO where I just, like, dump it all in there, and I get these nice, like, features. So, yeah, I don't really like and as do I like types? I don't know. I like, you know, gradually typing my stuff in PHP, but I don't like TypeScript.
Aaron
00:52:11 – 00:52:12
So you tell me.
Lane
00:52:14 – 00:52:27
Yeah. I don't I don't know. I guess I Yeah. I need to play with PHP a little bit more is is kind of what I'm, if I can have my own opinions. That's why I brought you on so I don't have to have the opinions.
Lane
00:52:27 – 00:52:27
You can have
Aaron
00:52:27 – 00:52:40
the opinions. Yeah. And my and my my great flaw or my great strength is that, like, I don't really have that many opinions that I feel like are applicable to other people. Like, I have a ton of opinions. I've got so many opinions.
Aaron
00:52:40 – 00:52:58
You could ask me anything, and I don't have an opinion on it. But I don't really feel like you know, I I have been put in this weird position of, like like, making videos and having followers on Twitter, and people expect me to say, like, what is right and what is wrong. And I'm like, I don't know. What are you trying to do? What are you trying to do?
Aaron
00:52:58 – 00:53:08
What do you like to do? What is your aesthetic taste? And, like, yeah. Well, maybe consider this then. And I just don't give, like, I don't give strict answers.
Aaron
00:53:08 – 00:53:32
Like, I I hate React personally because I feel like it's too low level and too finicky and cares me to like, requires me to care about things that I shouldn't as a human have to care about. The computer should have to care about it. But maybe most importantly, I find it aesthetically unpleasing. And, like, why why would I ever tell somebody you shouldn't use React because it's ugly? It's like, I'm not going to tell you your baby's ugly.
Aaron
00:53:32 – 00:53:41
Like, I'm just not going to use it because I don't like it. And so I get in this weird spot where people are like, What do you think about this thing? And I'm like, I don't know. Never used it. What's smelt?
Aaron
00:53:41 – 00:53:45
Don't know. Never never tried smelt. I have no idea. Why do you want me to have an answer
Lane
00:53:45 – 00:53:57
on that? You'll never tell someone you're not gonna date them because they're ugly, but, like, you're still not gonna date them, maybe. Right? Yes. I just saw that clip from it's, mostly tech.
Lane
00:53:57 – 00:54:01
Yeah. Right? Yeah. That was fantastic. I I loved that.
Lane
00:54:02 – 00:54:03
Okay.
Aaron
00:54:03 – 00:54:20
So that was that was a fun one. I was I was trying to, like, I started to use Remotion, which is a tool to build, like, videos programmatically, which freaking rules, and it's amazing. But it's a React thing. Ah. And so I used React I used React for the first time, and I was just like, y'all, I don't like it.
Aaron
00:54:20 – 00:54:31
I just don't like it. I don't wanna I don't wanna tell you what my dependencies are for some use memo or use callback or whatever the hell y'all are doing. Just figure it out yourself, your freaking computer. So, yeah, mostly technical.com. Check it out.
Lane
00:54:31 – 00:54:41
Yeah. Yeah. I I will be listening to more mostly technical episodes. Well, you had me at programmatically make videos and then, yeah, you lost me at React. We'll see.
Aaron
00:54:41 – 00:54:42
I know. Pity.
Lane
00:54:42 – 00:54:56
Might still have to check it out, but. Okay. So you're not so okay. Not a super huge fan of static typing, although there's some muddy definitions there. Do you get tired of typing dollar signs?
Aaron
00:54:58 – 00:55:25
You know, I have been doing it since I think I first picked up PHP when I was 13 or 14, so it's been 20 years now, and it just feels right to me, frankly. So all of our variables are preceded by a dollar sign, and people are always like, that is so weird. And it's like telling a fish that water is weird. I'm like, what are you guys talking about? About?
Aaron
00:55:25 – 00:55:32
I have literally no idea what you're talking about. Oh, the dollar sign. Yeah. That is kind of weird, I guess. No.
Aaron
00:55:32 – 00:55:39
I like it. I dig it. I think it's like I like it. It's like, yeah, it denotes that you're working with a variable. There's no var let const.
Aaron
00:55:39 – 00:55:43
It's just like dollar sign name. Yeah. Hell yeah. That rules. So,
Lane
00:55:44 – 00:55:53
that's my. Yeah. That that's my aesthetic argument against PHP. But again, it's not really something. It's not an argument you can take the bank.
Lane
00:55:53 – 00:56:00
I agree. Yeah. All I can do is silently believe that your baby is ugly. Yes. Exactly.
Aaron
00:56:00 – 00:56:03
And that's like, yeah. That's fine. I don't know. I don't
Lane
00:56:03 – 00:56:08
really care. That's fine. Cool. This has been awesome. Thank you so much for coming on.
Lane
00:56:08 – 00:56:13
I have one last question, which is just what's your spiciest web development take?
Aaron
00:56:17 – 00:57:08
I think we have passed peak React. I think I think React I think the React primacy days are behind us. I think a confluence of events, including interest rates going back up, companies folding, and also companies focusing on shipping you know, taking their goods to the market instead of tinkering with the vehicle. And a few, like, a few questionable choices from Vercel and the Next JS team, in terms of, like, feeling I feel like they're pushing complexity more than they need to, especially with this new, like, app router and React server components. They're, like, continuing to push complexity instead of productivity.
Aaron
00:57:09 – 00:57:30
I feel like people are starting to wake up a little bit and realize, you know, we are really spending a lot of cycles working on stuff and we're not getting a lot out of it. And so I personally think peak react everywhere all the time. I think that's behind us. I could be a 100% wrong, but I do think that we're getting close, if not there already.
Lane
00:57:32 – 00:57:36
React is a zero interest rate phenomenon. You heard it here That is first.
Aaron
00:57:36 – 00:57:38
That is my opinion fully. Yes.
Lane
00:57:38 – 00:57:45
I love it. Thanks again for coming on, man. Can you plug your stuff? Tell people where they can find all the cool stuff you're building and and making.
Aaron
00:57:46 – 00:57:50
Yeah. Absolutely. If you're into, alright. Buckle up. We got a list here.
Aaron
00:57:50 – 00:58:11
If you're into databases, planet scale.com/mysequel. I teach, like, an 8 hour course on, MySQL for application developers. Like, what do you as an app dev need to know about the database? We also have a good YouTube channel, in my opinion. If you want to learn how to do screencasts, screencasting.com, great domain name, great domain name.
Aaron
00:58:12 – 00:58:13
I'm going to be taking that course,
Lane
00:58:13 – 00:58:14
by the way.
Aaron
00:58:14 – 00:58:27
Yeah. Yeah. So that's a that's a solo outside project, screencasting.com. And then I spend a bunch of time on Twitter. So twitter.com/errandfrancis, that's where you can find me.
Lane
00:58:27 – 00:58:31
Amazing. K. Thanks again, Aaron. We'll talk to you later, man. Yeah.
Lane
00:58:31 – 00:58:32
Thanks for having me. See you.
Me

Thanks for reading! My name is Aaron and I write, make videos , and generally try really hard .

If you ever have any questions or want to chat, I'm always on Twitter.

You can find me on YouTube on my personal channel or the Try Hard Studios channel.

If you love podcasts, I got you covered. You can listen to me on Mostly Technical .