Hello everyone,

PHP comes up here from time to time, and I've noticed the discussion is usually based on what the language looked like 5+ years ago. Since I work with it every day and have genuinely come to enjoy it, I wrote a short article recapping where it actually stands today.

It covers modern deployment (FrankenPHP, Docker), software architecture (modular monoliths, the Symfony kernel, agents), the type system and its tooling (PHPStan, PHP CS Fixer), and the state of concurrency (ReactPHP, Swoole, the True Async RFC).

Full article: https://morice.live/posts/your-next-project-will-run-on-php/

Let me know if I missed anything, or if you'd like me to go deeper on a specific topic!

noticed the AMC theatres site has had queue times of over an hour today… just to get onto the homepage. That’s a bit strange right?

AMC has ~650 locations in the US. Assuming ~10 screens per location, ~5 showings per screen per day, and ~300 seats per auditorium (probably a generous estimate), that’s roughly 10 million available seats per day.

Even if site traffic is 5x higher than actual ticket sales, we’re still talking about something in the ballpark of 50 million daily visitors.

That’s obviously not nothing, but it also doesn’t seem like an absurd amount of traffic for a company this large. I’m curious what the technical/business rationale could be?

I work as a Developer Advocate helping people ship web apps as native mobile apps, and I kept answering the same questions over and over. So I wrote a guide that summarizes everything I recommend, based on 13+ years working in mobile development.

From what I've seen, every time someone in a web dev community asks "how do I publish my React/Vue/Angular app to the App Store?", the answers are either "just make a PWA" (which doesn't actually get you on the stores) or links to tutorials that assume you already know what a provisioning profile is and have Xcode configured on a Mac.

This one starts from zero. It's structured as an index, not a wall of text. Each step links to specific posts, videos, tools, and automation resources so you can go as deep as you need, at your own pace.

Some things I cover that I rarely see explained well:

- Why PWAs won't get you into the App Store or Google Play (and what to use instead)

- How to generate iOS certificates directly from your browser, without a Mac or Keychain Access

- The Google Play closed testing requirement that blindsides most first-time publishers: personal accounts created after November 2023 need 12 testers actively opted in for at least 14 consecutive days before you can go public.

- Why you should install the Live Update plugin in your very first release, even if you have nothing to update yet. Adding it later means another full native build and another review cycle, which is the last thing you want when you're trying to push a critical fix.

- How to build for iOS without owning a Mac

Honest feedback welcome: is there a step that's unclear, or something you think is missing?

https://capawesome.io/blog/11-steps-to-get-your-web-app-on-the-app-store/

I feel like this might start becoming a lot more common, due to all the hype and marketing surrounding AI. Seems a lot of ppl believe anyone can build an app in one-shot and we don’t need experienced engineers anymore.

When they ask a question like “why can’t we just one-shot the app”, it puts you on the hot seat to explain the entire development process to them, and all the steps it takes to get to a production quality app. AI can be used to assist and augment that process but you need a human in the loop to know and understand what they are doing. Rather than just telling them “go ahead and try it yourself”, how would you respond to this? How would you explain why AI can’t just make the entire app? How would you breakdown the step by step process of building an entire application?

Thanks

You can't imagine how bad my experience with Next.js has been recently.

I have two projects running on the same Ubuntu laptop:

• One Next.js app

• One TanStack app

The Next.js dev server was literally the biggest process on my entire machine, sitting at almost 4GB of RAM and absolutely murdering my old Lenovo. Even Brave and VScode consume less memory.

Meanwhile, the TanStack app was using around 800MB. Still not amazing, but nowhere near as insane.

Out of frustration, I asked an AI to help optimize the Next.js setup. It ended up changing some config to force Webpack instead of the default Turbopack setup and also added limits to how large the cache could grow.

Believe it or not, memory usage dropped from nearly 4GB down to around 1–2GB.

That's still a ridiculous amount of RAM for a dev server, but at least it no longer tries to consume every available resource on my laptop.

Maybe Vercel is thinking that everybody has a fancy Macbook M4 with 64GB ram?!

P.S. both codebases are small, max 50k lines in each.

I used to buy domains with godaddy but their pricing is deceptive.
They give domain for cheap on year 1 and from year 2 their markup is greater than 33% compared name cheap and cloudflare.
I am slowly transferring out all the domain from godaddy to namecheap and cloudflare

I've been specializing in frontend for more than ten years, gaining full stack and backend experience for the last four or five, and I'm starting to think I might just be done. Not to harp on the AI discussion that gets posted on here all the time, but I do feel like our jobs have fundamentally shifted over the last 12 months to something I no longer enjoy; I don't want to be a pseudo manager of AI agents, I want to write code, and if there's no longer a need for that I can just do something else.

But I would like this community's help - looking at it from the ground up, what are fully unrelated jobs or industries where you feel like our skills would be most transferable? And I mean fully blue sky approach, things I might not have considered, not just coding in a different stack. I think I have above average computer skills and awareness of operating system details that would put me above a lot of other candidates from different backgrounds, so where could I go to leverage that (that is at least a little more future proof)? Or beyond computer at all, I feel like I have spent years building strong critical thinking skills and analytical reasoning; I just don't have a wide base of knowledge outside our industry to know where these would be best used.

For practical purposes, I can't look at anything that would require its own degree and four more years back to school, but there have got to be lots of certifications I could get quickly to start to open doors and prove my qualifications. I'm certainly expecting a pay cut, but I'm open to entry level jobs if it means a good career path.

So I am interested in your thoughts - if you were starting out today in any other field than webdev, where would you be looking?

I won't name names because I got in trouble for that last post but I interviewed with the biggest agency I've ever gotten a call back from. They are 150+ employees, at least one major national corporation as a client.

You would think that because they are so high and mighty they have their s*** together but honestly no. The first interview was with the "big boss" the digital director. The second interview was actually 4 separate teams calls scheduled back to back with a total of 9 people.

The director loved me and told me I would be moving on. He said my second call would be with three people which is hilarious because he wasn't even close to correct.

My second interview was chaos. First call ended up being a different group of project managers than was listed, and they just hammered me with hypothetical situations of conflict the entire time. All rain clouds, no sun.

Second call was two developers who didn't seem that invested in talking to me. They mainly just said that works comes in from many different places and you can be expected to work on a bunch of things while managing a bunch of people AND be suddenly put in front of a client at any point.

My third call was hilarious because the top senior dev and the top senior designer in the company on this call loved me. We clicked on everything - our stance on AI, our views on WordPress and PHP in general, our approach to projects.

My fourth call was back to doom and gloom - more managers and all just hypothetical situations about conflict and everything going wrong. The people on call four were telling ME that a person from call two has a "different idea of what finished means" and were basically talking s*** on this person who is probably their smartest back end developer.

During all of this, the HR person had no idea who I was the entire time. I am interviewing virtually and would have to relocate and she sent me multiple emails about my "In Person Interview". My rejection email was actually the same email sent three times rapid fire from their HR software. I guess they really wanted me to know that are NOT interested /s

So the message I got was that HR is so overwhelmed with staffing they can't see straight. They will waste 10 people's time on a candidate and those 10 people will just project their workplace trauma onto you, or just use you as a nice break in the day to talk about something they care about personally. There are no processes, there are no boundaries, every project turns into a dumpster fire, and the smartest people in the company are treated like crap by management who don't understand the technical aspects of their job.

The job posting asked for 4 years of experience. They want someone who will work for 50k to 70k in an area where rent and utilities would be half of your salary but you are "passionate" aka young and exploitable.

I work for a nonprofit that’s had an outdated website for decades at this point. Upper management is kinda desperate and is getting quoted left and right.

$15K for a Wix site which includes: event management, volunteer management, shop, donor management, and general blogs, etc.

I thought Wix was one of the lower quality sites… especially as I can just go in and drag and drop elements myself.

Are we being highballed? How can I convince my management who has zero website experience that this is not the route we want to go for that price point?

EDIT: wow y’all really came for blood in here. I mean no disrespect when I say “click of a button” and “drag and drop,” as I know web dev is not easy and requires tremendous skill and knowledge. All I’m saying is if we’re going to be paying $15K+ (as a small nonprofit with over 10K database) then a site like Wix just isn’t acceptable? I know y’all agree with me! I appreciate all of y’all’s knowledge and advice.

O'

Billion dollar companies. It boggles the mind.

Probably takes $500K in tokens for AI to figure it out.

</rant>

I've been doing webdev for over a decade. Now I feel burnt out and worried about prospects. Mid-career and barely getting interviews.

Not only do we have to compete on a global stage (I'm US based), but we have to deal with regular layoffs (and that's accelerated due to AI).

What used to be reliable work with good pay, seems to have fallen off dramatically in both aspects.

Do you agree? Would your recommend this field to a friend?

I've been building out embedded reporting for b2b clients (i use next.js), and as our user base grows, the P95 latency is becoming a nightmare. Every time a user changes a date filter, it triggers a fresh compute spin-up on the warehouse.

I’m currently refactoring the stack to put a universal semantic layer (in my case, cube core) between the frontend and the database. and the goal is to move all the logic, like multi-tenant row-level security (RLS) and complex joins, out of the React code itself and into a declarative modeling layer

since I started, the biggest win (so far) is using pre-aggregations. now, instead of hitting raw tables, the API hits a warmed caching tier which is in the 'cubestore'. it feels more like querying a structured API than a database

for those of you here who do high-concurrency analytics in SaaS, a question! are you just throwing more money at warehouse compute, are you moving toward this kind of decoupled architecture, what other best practices do you have? now trying to figure out if I should stick with this or just move everything to a huge ClickHouse instance and hope for the best

I need to share my misery with someone. Hope you get a laugh out of this.

About a year ago, I built a web app for a client. Let's call her Karen. She loved the UI; she loved the functionality - every meeting was a joy. She was shit-kickin happy. I have 10 months of emails from her saying stuff like "it's like you're in my head! You get it exactly!" She even paid me extra for new functionality that she came up with halfway through.

Then, after I had delivered everything she asked for (and signed off on), I demonstrated the finished app to her and her staff, and I thought it went well. I kept asking her when she wanted to launch it - crickets. Then she called me two weeks later, saying that one of her employees was still using her old system and asking Karen why. Apparently, the employee said that what we built didn't meet her needs... no details. So then Karen lays into me and says that what I built her is worthless and we need to start over. This is just out of the blue; absolutely no complaints until then. She was literally screaming on the phone. My wife heard this because I put her on speakerphone. I told Karen, " Hey, I'm sorry, but you have never said you weren't happy or that anything was wrong. I can't start over. I have to pay my staff to start over. If we did something wrong, I would cover the cost - but I built what you asked for, and I have many emails and Zoom calls recorded where you were happy."

Then I don't hear from her for about four months, and she sent me this nasty-ass email saying that I screwed her over, used templates off the web (not true), and she wanted $45k in compensation (more than she paid) - or - make up for it by redesigning her Claude designs for her other stuff. "I did it on the weekend in 2 hours, I don't know why you developers charge so much!"

She would never win a lawsuit. I forwarded everything to my attorney - and he laughed. He said, "No way she can build a case. But try to settle with her and do what she asks; it's not worth the fight."

So me, being the OCD dumbass I am, I not only redesigned it - I built the entire application, too - 11 modules, stuff like an employment board, full listing site, and much more. Probably what I would have charged 200k for before AI. I have been doing this shit for 30 years, and I guess I was having fun. But it was like "I'll show you!"

So then I say I got all this done, and I have a big surprise for you. So we get on the call, and I show her that I not only redesigned all her stuff but also made it live with logins and functionality for everything - not just design - backend, everything. I'm telling you this was polished. I looked over it by hand, fixed stuff, tested it, ready to go. I thought she would be impressed and happy, and she seemed to be... then... crickets. For months.

Finally, I called her, and I said, " Hey, when are we going to launch this? And she replied, "Launch what? The Claude templates you coded and turned brown?" Then it dawned on me... she didn't get it. She didn't understand what I did. Hundreds of thousands of lines of code. I told her. I showed her.

Then I shot an email to her new developer, and he gets back to me, and he says, "holy shit, this is awesome! And he somehow translates this to her. Finally, he sends me an email and tells me that he recommends that she launch with what I built!

So, it's all over, almost two years later. I swear... I don't stutter. I've dealt with clients for decades. I know how to explain technical stuff to people. I guess she has ADD or something?

I feel stupid for going above and beyond. I have always done that. And clients don't get it, and they don't give a shit. I am so sick of it. But it's also pretty hilarious.

I'm a Full Stack dev. At my current job, until recently, I've being stuck doing only front end work. This is due to coworkers (referred to as C:A & C:B) hording all the back end work. They have stated that they don't know the front end. Yet they use A.I. to generate front end code they don't understand. Tech Lead doesn't care as long as it works.

To be clear, this is not an anti-A.I. post. I see it as a tool and use it myself. Whatever it generates, I go through and understand how it works and if it is the best way to do it.

When I say "Done Dirty", I mean A.I. is being used by people to do front end work that they wouldn't be capable of otherwise. Examples of what I have experienced.

I was working on an assignment with Coworker C:A. I was on the front end work and they on the back. C:A branched off my Git branch, used A.I. to finish my part and theirs. They presented it to the boss (a non-technical director) as their work alone.

I was working on an assignment with both C:A and C:B and we got down the wire on it. It was the day before delivery and I was trying to wire up the U.I. for both their back ends. I was running behind. C:B decided to use A.I. to wire theirs up and told everyone they got it done. The next day I was looking over everything and realized it wasn't wired up. It was displaying the embedded data I put in for testing. I had to inform the Tech Lead and it did not go well for C:B.

My question is are front end devs getting the "short end of the stick" because A.I. can generate U.I. elements and do interface work for people who don't understand anything about it. Is anyone else experiencing something similar? Not looking for anti-A.I talk. I really want to know if people who specialize in this work are being directly undercut by A.I. usage.

TLDR: Are devs using A.I. to generate front end code they don't understand undercutting those that specialize in it?

I've been doing frontend development for a little over a year. Mostly personal projects, a few freelance jobs for friends, nothing huge. Recently someone reached out through a mutual contact and asked if I could build a website for their business. At first I assumed it would be a simple brochure site, but after our call it sounds like they want something a bit more custom.

The site itself doesn't sound overly complicated, but they have a bunch of specific requirements that don't really fit into an off-the-shelf template. The deadline is around two weeks. Part of me thinks this is exactly the kind of project I need if I want to improve and start taking on more freelance work. The other part of me is looking at the timeline and thinking there's a good chance I'm being way too optimistic.

For those of you who freelance or run your own projects, how do you figure out whether a client request is genuinely manageable or whether you're setting yourself up for a stressful couple of weeks? Trying to work out if I'm overthinking this or if my instincts are trying to warn me about something.