My PHP Problems
dantleech.com
dantleech.com
Going from zero to functional web backend in less than 3 hours for a language I'm almost totally unfamiliar with made me understand just why this got so gosh darn popular. I thought building out a Flask application was fast, but PHP's initial experimentation loop is just on another level, criticisms be what they may.
But I regularly write fast, robust, secure backends, in a day or less (less, if possible. I don't actually like working in PHP).
Despite the torrents of hate from this community, it is still pretty much the backend king.
Here's the "Fishtank Graph."[0] It kind of puts things in perspective.
[0] https://w3techs.com/technologies/history_overview/programmin...
However, I would take the numbers on the site you linked with a big grain of salt. Detecting which backend technology powers a site without direct evidence is nearly impossible. PHP may be high up the list since it and its associated stack leak the most metadata.
(Not really but it makes me giggle at least)
Not only that, but since .NET Core and the new versions of .NET running just fine on Linux distros, I'd expect their usage to go up, not down.
Curious. Actually, their disclaimer page seems to suggest that setups with a decoupled SPA/API might not be detected correctly: https://w3techs.com/disclaimer
I've spent 15 years writing serious code and never ran into Python all that much until I went back to engineering school. I don't think a lot of people use it for their backend code, at least in my experience. But wow are there lots of stats and science people using it exclusively--it's completely ubiquitous in academia and research (though R is making headway).
Think about how many Wordpress installs there are, not to mention Wiki's and forums.
> It is powerful indeed. But PHP syntax feels unelegant and making me write unnecessary ceremony. It feels different when I write typescript code, every character I type fits like lego.
> Especially when my codebase grows so much. Navigating PHP project is like thrown in forest without a map, you can always look up the sun to find your way.
“Elegance” is something that gives people warm fuzzies, I think this is why there are so many Rust advocates (I am one), when pieces fit together nice or can be refactored cleanly and tests are first-party it is nice.
However there have always been “hacky” and “proper” languages, because doing something hacky that works also feels great.
Python is a hacky language, it’s easy to go to 0-100 very quickly. C++ however requires time and precision to make things proper, so people avoid it and few things are made comparatively in C++ unless it's infrastructure software that people spent a lot of time writing.
Maintainability be damned, sometimes you really just need to throw something together.
The curse of things that work is that they will be expanded upon and kept around for too long.
Unfortunately: PHP works, and is an excellent hack-language for the web. Which means there’s a lot of code that becomes “production grade” that should really have never been intended to live forever in the first place.
I'm working on my own startup that started as a side project and it was super hacky at the beginning, but by sticking to PHP (and thanks to PHP continuous improvements) I was able to refactor little by little and make the codebase way more structured and easier to maintain.
(I keep wishing that I could rewrite it in a “better” language, but there is no other language that is even in the same ballpark when it comes to “any random user can upload this .php file to any random web host and expect it to work”)
This is a trade-off that makes sense to embrace in some contexts, and I do believe that web is one of them. The more nicely your pieces fit together the harder it'll be to break them apart. You can only refactor cleanly once you are aware of all the requirements, but those requirements are usually constantly changing in the context of web and so your pieces fitting together a little too nicely means that you now have to refactor a lot more code.
Go is the only other programming language I've had this experience with - picking it up with a goal in mind and succeeding in doing something more quickly than I might have with a familiar language.
That’s what it started as.
It’s entire process model (instantiating the world for each request then tearing it all back down, not having any shared state) is fairly unlike most other general purpose languages.
PHP inverts the usual assumption that “this code file is primarily code” on its head—it assumes any file it’s given contains other data until it finds the special open/close tags signifying a section that contains code. (Allowing you to embed it within HTML easily.)
The runtime handles things like GET/POST variables natively and has a special way to expose them. Ditto for file uploads.
The standard library isn’t extensive but contains a ton of functionality that revolves around managing web requests. (Managing HTTP headers, setting and reading cookies, ...)
Etc, etc.
Especially now with generics support. I’ve been able to quickly write little meta functions for things like translating Kafka events to a channel of structure and spin up a number of go routines to handle them by making REST calls. The code is concise, clean, and simple, without being overly clever.
Of course this is not what we see at production-grade software, but I really enjoy the fact that the language was designed with the HTTP protocol in mind. It's really unique in this sense.
When I did learn about HTTP, it was very easy, since I already knew it without knowing I knew it, so maybe that's in favor of your point, but there's much to be said for the actual understanding that I didn't have at first. When I started interviewing people for PHP entry-level jobs, asking about HTTP was one of the ways I gauged how well applicants understood their work at a conceptual level.
It also took learning other languages to understand how PHP is so unique in how tied to HTTP it is (rebuilding its entire universe every request).
(these learnings happened when moving beyond the PHP-based CMS style of development - agency style work on my case - to make more custom software).
The fact that I can't tell if this comment is sarcasm or snark or serious says a lot.
PHP mixed with HTML was one of the "bad ideas" that hurt PHP. It was a bad practice that got out of hand.
Now why no one bats an eye at node/react doing this very thing and calling it JSX is beyond me but...
Hey don't drag Node into this, you aren't forced to do any of that.
React to me feels like SPA developers discovering the idea of PHP and thinking they invented it. This shouldn't be surprising, since React came from Facebook, and Facebook historically is a PHP shop.
Mixing your business logic and presentation is a bad idea, but it's very attractive when you don't have the informative scars
You might say, doesn't Facebook have the scars to avoid designing React? Well, when you have a hammer, everything looks like a nail.
PS: I would suspect the down voting is just the "LOL", otherwise it's a fair comment.
This is accurate, and I never thought of it that way.
> PS: I would suspect the down voting is just the "LOL", otherwise it's a fair comment.
I did laugh out loud because I did not know WHAT to make of the comment. As someone who buttered my bread on PHP when everyone hated it, it was a sign of how times have changed!
It works great, it's fast and you don't have to learn a new templating language or dialect every 2-3 years. Or try to figure out how to do trivial things like loops in loops.
I don't quite get the use of templating engines in PHP because PHP is a templating engine itself.
[1] It's so much worse than this... You can use some horrible hacks like undefining a well-known function that each template implements, but this would only work one level deep without also adding a mechanism for "calling" a template, and at that point you are making a template system on top of PHP.
$_SERVER, $_POST, $_GET.
The rest is not the PHP3 I recognize to be honest, let alone the modern stuff. Remember include() literally just included the code right there, inline, so you could include in functions and gain the function's scope:
function render($data) {
switch ($data['foo']) {
case 'bar': include('bar.inc'); break;
case 'baz': include('baz.inc'); break;
}
}
> Remember include() literally just included the code right there, inline, so you could include in functions and gain the function's scope:
I've talked about this approach elsewhere in thread, it is not nestable without building a way to "call" templates on top of require/include, and when you do that you'll need to undefined the render function to do it as well, which creates additional complexity. At that point, the original premise of just using raw PHP isn't very true, because you are now building a framework or templating system around it.
Also, you have to contend with the magic "render" function not being globally defined by something. It's messy.
Um. Can you give me an example to pad that out a bit?
> and requires you to do cleanup after each template instantiation to avoid leaking values between templates
In my render() example, isn't that "reaching the end of the function"? (Edit: Wait, you're only talking about the superglobals, right?)
Your example of using a render() function is already an example of this. You can parameterize render() to provide data to the template from the caller without utilizing untyped globals, and without having to clean the globals up after the template has rendered.
> Wait, you're only talking about the superglobals, right?
Yes. Globals are globals. PHP's super global distinction isn't really relevant to whether you are tainting the execution environment when you pass data to a script.
The point is, as I elaborate elsewhere, PHP is very awkward to use as a proper templating system.
I hear this here, and elsewhere I hear that PHP has evolved into a decent language. Both can't be true.
Namespaces allow non-global functions without a wrapping class already, IIRC. Looking up though, there seem to be special cases (legacy behavior, making everything at the top level global) when using namespaced files with include() statemtents, if I read the docs correctly.
Dumping pure functions as static members into wrapper classes was not uncommon when I used PHP.
But primarily, a lot of code was heavy on dependency injection and inversion of control, Symfony being the leader of this coding style maybe.
TFA also seems to be about class-oriented programming, but not the heavy design-pattern style mentioned above.
The linked TUI CLI parser looks like a nice and lean PHP OOP example.
$closure = function(){};
instead of
function closure(){}
1. It's a loaded foot gun. PHP was notorious for 1000 line files with HTML and SQL all mixed together. (I speak from personal experiences on this)
2. Templating languages are "portable" (In theory). Have a mixed env of PHP and Java... there were templating languages that would work for both teams, it took a step out of scalability.... XML and XSLT were the promise land on this one, but we fucked all that up. Portability is a big deal. We're still slinging CSV files for a good reason.
Improper use of templating engines or trying to use templating engines that aren't up to the task can give you headaches just as well. Sadly no amount of frameworks or templating engines can stop a bad programmer from writing bad code, in the end we're craftsmen who need to learn how to use our tools.
2. This could be a valid use case, but a rather rare one. XML+XSLT is something that sounds fantastic on paper, but as anyone who actually worked with it knows it ranges from a big disappointment to an absolute nightmare.
It's not a matter of whether the code is bad or not, but how bad.
The entire point of frameworks is to add guard rails to help you stay disciplined and avoid cutting corners on separation of concerns.
Using PHP as a template language tempts developers to violate SOC every day.
Also I don't see how using a loop or conditional in PHP is any different from using one in a templating engine, aside from the overhead of the engine and added annoyance of debugging another language in your project.
Those problems are called "junior developers", and I think we'd all like to avoid them, but that's not very sustainable.
Perhaps we can fix the education problem, but given that folks just do a boot camp and head into the market, prospects aren't looking good.
> Also I don't see how using a loop or conditional in PHP is any different from using one in a templating engine, aside from the overhead of the engine and added annoyance of debugging another language in your project.
The problem is convention, or lack thereof. Without convention, who's to say where that database call should live? Frameworks, which tend to use templating languages for this reason, tend to make that convention clear to all teammates.
I want to mention that this doesn't mean you can't use PHP as the programming language for your templates as long as the conventions are clear, but PHP is poorly suited to being a template language for reasons I've posted elsewhere in this thread.
Its a one off... Its just an internal tool... This is a prototype... It's a hot fix...
I know better, I have done it. It is only with dumb luck that they didn't end up as products. Some people dont know that they should not or that it isnt a place they should try to "get away" with it. These things grow and then turn into products and someone has to clean them up.
Source: I made a lot of money turning PHP "projects" into "software".
> Also I don't see how using a loop or conditional in PHP is any different from using one in a templating engine
One of these things is Turing complete. The other is not.
Templating languages have their own laments: see yaml configs used to create "workflows", "environments" and "servers" via templating.
At another job, another guy built some massive abstract-factory-factory-layer in Python, a class hierarchy 4 or 5 layers deep. 5000+ lines in a single file. It was almost impossible to debug.
Both of these are better than some C++ code I once saw. 10,000+ lines in a single file. In that file, there was a function with a 3000+ line case statement, which was mostly copy-pasta'd from other parts of the code (not in that 10K line file.)
I have seen plenty of PHP code that smelled like a nuclear dumpster fire, but you can write trash in any language.
Do you write safe templates in PHP ?
In PHP, “HTML" is string concatenation; in React, “HTML" is an HTML-flavoured bit of syntax sugar over HTML-generating functions (and you are still welcome to call the functions yourself, if you want your HTML-generation code to avoid looking like HTML). Aside from looking kind-of similar if you squint, I’d say the differences and pros and cons are fairly significant (For example the IMO worst thing about the PHP approach, the ease of shooting yourself in the foot with XSS holes, is a non-issue in JSX).
I use it exactly for rapid dev of stupid ideas. I’m not sure if I’d use it in a production environment because it is quite slow.
Indeed.
It is obviously good to be able to automate generating web pages based on say the state of a database. As this gives you up-to-date pages without someone having to manually build them.
And if that - automatically build web-pages from a database - is your goal and your current state of the art is static pages, then creating a language you can interject into your static pages, to build the page dynamically is a no brainer.
So what, JSPs are just the same. In the end what matters is the speed to Get Things Done, and there's utterly nothing competing with PHP, closely followed by Tomcat+JSP as you can use the insane amount of interfaces with just about everything that the Java world has.
PHP receives its share of criticism, and HTML lives on relatively unscathed.
So if mixing PHP/HTML is a problem, you'd probably throw out the PHP and keep the HTML.
Also, isn't this use case PHP's origin? A dev wanted some programmable HTML?
Because it's not "this very thing". HTML around PHP is output directly to stdout. JSX is just syntax for literals.
If you think of PHP as prototype-only, mixing tags and HTML is very convenient, like it is with the other thousand templating tools. And obviously it's more a problem of XSS through badly sanitized inputs more than the mix in itself, blah blah blah
It's PHP in production a bad practice that got out of hand.
Turns out I wasn't missing much. I've mentally pegged it as an example of "the pendulum swings" - thin client vs thick client, centre vs edge, tight coupling vs loose coupling, etc.
React is popular for the same reason as PHP. I can only hope the industry corrects for this soon, because I challenge you to find a startup building raw PHP in 2024, yet nearly all of them build with React.
Especially when my codebase grows so much. Navigating PHP project is like thrown in forest without a map, you can always look up the sun to find your way.
A breath of fresh air in amids of thousands of "ew, such a bad language" and "a fractal of bad design", the latter being incredibly outdated and still referenced from time to time.
I love PHP, it was my first language and it's one hell of a productive ecosystem to be in. Laravel is an excellent choice for rapid prototyping and Symfony is probably one of the best web frameworks for enterprise projects. Paired with coroutines extensions such as Swoole/OpenSwoole/Swow it puts one hell of a fight performance wise with more modern languages.
Also, I have been looking for GraphQL implementations for other server-side languages recently and I have yet to encounter a schema-first implementation that comes close to what the Lighthouse layer for Laravel can do.
That is really productive.
And it gives me the opportunity to separate out the front end concerns in a way I could explain to someone else.
Apollo has been a bit of a pickle on the front-end, mind you. That's a decision I'd revisit.
But generally it has meant a flexible API interface and leads to nice low latency UIs.
Check it out:
And these really good third-party auth mutations for Passport:
And I _hate_ Eloquent, but I'm that jealous of all the goodies in Lighthouse's graphql dialect. You pretty much can't use directives at all in API Platform :(
I'd only seen fairly raw GraphQL from like 4-5 years ago, when you had to write your own auth, resolvers, deal with n+1 queries, etc. This is actually really nice and solves a lot of the issues I was thinking of.
Schema-first feels more manageable.
I'm not going to say I haven't had to do that a _bit_ here, when I've got into quite advanced things.
(e.g. sometimes you might want to dynamically provide the values from an enum, or write your own decorators, or you might want a custom top-level resolver.)
It's built on the WebOnyx reference implementation, so there's some help out there.
But it's amazing how much code I haven't had to write, and being the sort of person I am with difficulties controlling focus, I absolutely love the schema-first approach. It interacts well with Eloquent scopes, with the ordinary API gate system, with policies and validations.
It's actually weird how good it is. You still have all the same issues you have with GraphQL generally (like limiting graph traversals to things a user is authorized to see), but you have so many tools.
Working on FrankenPHP, we get a lot of issues from the Laravel people (mostly in regards to octane + worker mode), and I've been getting into more and more Laravel to support these folks and understand what is going on in the SAPI. I've really started to understand the attraction lately. Previously, I worked at Automattic for quite a while and learned the WordPress side of PHP, then worked with Symfony in a corporate setting, so I've yet to get into Laravel in the commercial world, but I'm thinking I probably will for my next job.
I've still to find a better code first implementation too
I agree that the language evolved since then, but the main categories are still not solved in the language. Nice frameworks just hide those well.
PHP has improved since in a way that is rarely considered when reposting it, but JS has some deep language horrors of its own still, and many novel JS horrors like NPM have emerged largely since the Fractal article was published.
As the King sang, so eloquently: o/~ clean up your own backyard o/~
PHP is no longer a terrible language- recent editions have made leaps and bounds on language features, and the ecosystem around Symfony is great. But on a weekly or monthly basis I run into something that makes me want to head-desk.
JS's recent(ish) language improvements like `let` and ESM have been much more fundamentally significant than PHP's IMO, where we're still stuck with bad scoping and Java-brained class autoloading.
*Not leagues apart in terms of productivity. I can certainly get a lot done with PHP. I mean leagues apart at their foundation, the basic concepts at work in the language. Most of that can be ignored until it rears up to harass you. Here's a great example of something similar to an issue that bit me last month:
As someone who’s written a lot of php and JavaScript, I think the same thing but in the opposite way. JavaScript just can’t compare when it comes to designing a complex OO system.
Adding properties to an object is fine IMO. Arbitrary properties and prototypes are, to me, a much cleaner system than PHP's somewhat confused mix of array, object and class. The lower you go in JS the more unity of concepts there is, whereas the lower you go in PHP the more hacks you discover.
All of the above said, I would prefer a string concatenation operator to overloading +. JS template literals mostly solve that for me in practise. And what I really actually want is an equivalent of Haskell's <> or mconcat, which concatenates anything monoidal - strings, lists, etc.
Rather, point out the fact that Symfony is a framework that relies entirely on all sorts of magic behavior, and at the same time it is entirely undocumented.
One of the most horrible programming experiences I've ever experienced.
PHP is a nice language, but the community is insane.
The framework has changed massively after the earliest versions (1~3). These days it's very similar to Java's SpringBoot. After scaffolding all you get is a thin layer with the HTTP abstraction, configuration support, reflection-based Dependency Injection and a testing framework (PHPUnit).
Since the framework is component-based you pull dependencies as you need them. I find the documentation really good and honestly I don't see much magic in it, it's an event-based web framework. Laravel and it's facades on the other hand is very magical.
Where are those goto statements? I'm asking out of curiosity to check them.
https://github.com/search?q=repo%3Asymfony%2Fsymfony+goto&ty...
I highly recommend learning how to use it.
That's not a positive. Spring Boot is impossible to use for anything except an unmaintainable abomination, since the whole thing runs on undocumented magic.
You have the docs with extensive examples, you have a free book, and the whole framework is typed and all configuration options and the individual components referenced: https://symfony.com/doc/current/index.html
And if you use PhpStorm with the Symfony plugin almost everything gets intellisense.
Also, random code snippets that may or may not work is not documentation, much less is it a reference.
(No, random posts oт Stackoverflow and generative AI responses aren't a replacement for documentation either.)
The blog author makes solid points that I agree. Meanwhile AFOBD just make fun of stdlib inconsistencies and of language quirks, most of which were fixed throughout the years.
AFOBD is so incredibly outdated that this paragraph is still there: "PHP is naturally tied to Apache. Running it separately, or with any other webserver, requires just as much mucking around (possibly more) as deploying any other language.".
It's now a powerful, fast, flexible platform with a load of amazing libraries and tools -- graphql, webhooks, databases, all sorts. There's an amazing level of recent PHP support out there you can host on.
And then there are the terrible applications, written from inadequate specs, in a hurry, by people who just aren't developers. And they are embedded, and people want them fixed.
You have to have a knack for, and some enjoyment from, that work.
The worst I have encountered is a Wordpress site built by some people who did not know Wordpress could do several important things, so they'd subverted the loop, ignored where they could hook, and embedded wordpress outside the loop to do stuff, which was a nightmare.
The second worst was, it must be said, a Laravel app, but that problem wasn't the PHP so much (though it had been written in a hurry with an awful lot of duplication that we could have refactored out), it was the unreliable assumptions made about MySQL default result ordering.
Which is actually the secondary problem here: people who make bad PHP decisions usually make bad SQL, email and HTTP decisions too.
If building your own thing from scratch, or starting a new project, etc., PHP is a blast now, actually. But the very old things... not so much.
Of course that was never guaranteed and subsequent optimizations and changes have really removed that ability to ignore order in many situations. Plus, MySQL also changed to actually check if you're doing GROUP BY properly by default.
MySQL was always a big part of the PHP story so it's fitting they've followed similar arcs of improvement and growing pains over the years.
at some point in the past PHP was the most productive tool for some quick & dirty coding, but not anymore for me.
I’m a contributor over there.
Wordpress is kind of it's own ecosystem. I have been a PHP dev for years and never touched any Wordpress code.
Most jobs recruit for specific frameworks like symfony or laravel.
Modern PHP is great, the past & some fundamental decisions are automatic tech debt that i can live with. It'd be great if PHP 8.3 gets forked with no backward compatibility, removing all the old PHP as a template engine designed functions that nobody needs & release under a new name. I just can't find the same type of joy in writing code as i do when writing PHP. But some things are truly annoying. Just not enough for me to switch or start the fork myself.
> but you're left with an incompatible language in the end, at which point you're probably better off starting with a different language entirely.
Yes, but which one?
No matter which one i try, in the end, i am always close to rewriting it with php. Eg as many good things go has, i don't enjoy the syntax as much as phps. I could live with it, given the advantages. But i can't get over it, that most of my code seems to be error checking. Sure, that's probably a good thing for serious work. But that's just not where the joy in programming lies for me. Most cases phps try-catch feels much, much more elegant / fun to write (not comparing the efficiency of the outcome).
I do love the fact i end up with a single executable. And php without framework requires much more overhead, which is also annoying. And i am done with php frameworks.
Obviously, PHP came from a place of great inconsistency, and the remains are still there, including the popularity to rant on it. PHP is getting significantly better though, and while a long one, I hugely prefer this route over something more abrupt like python 2-3. The first 3 or so points are valid, but annoyances at best. For the points about closures, statements, inline classes etc., they are just features from other languages that aren't in PHP.
The "living on a prayer" comment on serialization brought back memories. It made me acutely aware that I use this pattern in go, and there it "just works". I now have a note to go see what Valinor, PHPStan, Psalm, Serde are because they sound godlike.
Great disclaimer, since most of the points raised in this article seem very nitpicky to me. This is understandable considering that OP's PHP work is about creating a language server and some benchmarking tool, both of which are definitely not the main use-case for this language. I bet your average PHP dev has no idea about the things mentioned in this article, nor are they interested in improving them.
I hope that discussions like these raise awareness within the PHP community and eventually lead to solutions. Considering the impressive initiatives the PHP foundation has recently undertaken, I am confident that the language is evolving in the right direction.
Certainly, there are still individuals in the PHP world who may not grasp the essence of this post. In some tech interviews, I've encountered people stating that they 'don't know PHP, only Laravel.' I don't oppose Laravel (or WordPress or any other framework); I just want to emphasize that some individuals may not appreciate the underlying technology. Reflecting on my early days of programming, there was a tendency to exclusively focus on learning jQuery, neglecting the importance of understanding JavaScript.
Same goes for container data types. All you've got is array. No sets, no hashmaps, ordered sets, sorted sets. Array is just all of this at the same time. There's also an extension for that, but its API is... not optimal. I wonder how much faster some programs could be if the array wasn't used as the universal data type.
If one could use composer to install PHP extensions, that would be awesome. I fiddled around with WASM as a language for universal PHP extensions that can be loaded via composer, but there does not seem to be a lot yet.
> I love the smell of false in the morning, but despite that I do wish that all PHP’s built-in functions threw exceptions. We have the famous safe library which does just that! But if you’re like me then you don’t like coupling huge amounts of code in perpetuity to an external library.
Although it does require some boilerplate, which may seem like pointless extra work.
> [...] only to realise there the class has a private constructor or that there are static constructors which I should be using [...]
Oh but they do. You should spend some time with those languages so you can do a proper comparison.
Go has had new for a while now.
In sum, the native functions should take iterators anywhere it can take arrays.
Every other valid criticism he makes and many more I can wrap in the simple statement that: psalm needs to become the runtime type system which is hopefully and appears to be the (maybe distant) future.
In the meantime you have 40% and quickly growing of those features in your IDE and 100% in your CI so it’s not a huge deal.
Main takeaway for people who haven’t looked at PHP in 10-15 years or since the fractal article and would take the length of this post, which they don’t understand, to say “yep dumb php again” no.
—————
Constructors Only had this confusion once with my UUID library and it took me 3 seconds to remember the formula.
"In attribute land this becomes:" I've never seen any code or library that uses attributes like this nor do I understand why you would. The only purpose would be static analysis, let's say with psalm, and that's just not how you use attributes in psalm. He says generics are tricky: in psalm they're trivial. The real problem is the best IDEs are still working on getting the integration right.
Nested attributes strikes me as an extreme edge case that would only come up in some meta-level tooling you're making for the language which is the provided example. However in this special case presumably this can be accomplished through an extension, xdebug for example does this.
Serialization he says is a problem then says, it's a solved problem (there are many such accepted ways to solve this) but then says it still annoys him so not sure?
"No Variadic Promoted Properties" I'd like to hear the use case for. I've done very complex things in PHP and never needed this construction. A property can't have a static single type but then also consist of a collection of random types, in this case it would just be an array. It looks like another feature that is solved by psalm.
Iterator to Array: Preserve Keys he says he doesn't know if it's wrong.
Iterators vs. Arrays this is easily fixed by using the popular Functional library versions of these functions which I agree should be the default.
Short closures cannot have statements" I have to convert between them so I kind of get it but would say this is an intentional choice. You can substitute a method call if you really want this.
"Statement Blocks in General" That would be nice when and if it's very simple though having them anywhere could turn the language into kind of a mess to read considering what those blocks could contain.
Functions that return false - basically everyone agrees on this, it's a conversion that's iterated on with each release and will just take time. The specific example he provides I have wrapped with an exception thrower for just this reason and never use the default.
Inline Classes - he says he doesn't know if it's a good idea. I'd say no, having a reliable file system convention between codebases is a plus not a minus.
That said, I'll comment on the code style this author wants to use. It's flashy, confusing, and unnecessary. This person wants to show proficiency with the language by using all its features. That's not how you write good, readable code. Come up with some personal standards for yourself and stick to them. Stop mixing every possible style. You're increasing cognitive load.
Continuing on, you want to use all the latest features of the language. That's great, but why are you trying to skin the cat three different ways in the same program? Pick one way and forget the others. If you can't fit all your code into short closures the rest will look like shit and confuse everyone else. I see this all the time with JS guys mixing bracket and dot notation. I feel like there is probably medication that can help with that.
Then we get to the fact that you want everything to throw exceptions. Like file_get_contents. No. Just no. You're not supposed to rely on that function to tell you if the file is there. That's what file_exists does. If you've made it to file_get_contents and you don't know if the file is even there, you're doing PHP wrong. You can use that function with a URL and eval to execute remote code. You're not supposed to swing it around like a broken bottle.