|
Call me an old goat but I disagree.
I am not scared to learn new things, can be seen my Q&A post, fairly well written in some new tech but I still swear to PHP. If you kept track of their new technology, you will be surprised how well it does currently if not out performing some newish languages.
If Google is anything to go by, it still lists in the top 10 of all languages that is preferred - where does PHP 8 rank as a language[^]
We can then rant on and on about the pro's and con's but IMHO it sis till ranking up there for a reason. I am busy with an article to be posted soon in PHP as we still get so many questions/references to PHP that I think it still deserves it's seat on the counsel.
Ok, pull the big guns now...
|
|
|
|
|
You ever do a custom build of PHP or heck even a custom install on a server you're administrating? Or the pain of configuring it with Apache, etc.?
I dare you do a Node/Express project just once. I dare you.
Jeremy Falcon
|
|
|
|
|
For me NodeJS or even Python was never a solution to replace PHP. First I would go for TypeScript instead of JS and Bun instead of Node/V8. Take a look at Bun — A fast all-in-one JavaScript runtime[^], you can run TypeScript with it directly (though it is internally converting it to JS not running TS directly). Probably a good replacement for PHP was Scala IMHO, not js, but that was in the past, now Lightbend prioritised Java support and that was a disaster for Scala.
I personally now use Haxe to target both PHP and JS runtimes, Haxe is stronger typed than TypeScript and that's why I prefer it, at the same time I can target C# and Java runtimes from the same Haxe codebase (this is not possible with TypeScript yet). But unfortunatelly Haxe doesn't have php8 support yet. The advancements in php8 are stunning. You should take a look.
|
|
|
|
|
TypeScript is a superset of JavaScript. You still run "TypeScript" code on Node after it's transpiled. Also, I've looked at Bun, but it's not as mature as Node. Maybe one day, but not today.
Jeremy Falcon
|
|
|
|
|
TypeScript is a superset of JavaScript.
Correct, but it is still an improvement that JavaScrip neeeded, in my opinion. Of course, not everybody agrees or needs to agree.
The whole point is that I could never switch to Node for backend (even though I've tried) because I found JavaScript too dynamic, even though I was writing JavaScript for the browser, but at that point I had no alternative, WASM to the rescue.
|
|
|
|
|
You're missing the point actually. You can do TS in Node. It's not as "dynamic" as you say. I guess you just wanna argue. Tootles.
Jeremy Falcon
|
|
|
|
|
No, I dont want to argue. Of course I can do TS in Node, but first TS was widely adopted later on than Node as far as I remember. And as I already said, I can equally use Haxe to generate JavaScript for Node as I can also use Haxe to generate PHP.
I just think ... why should I switch from PHP exactly to Node when I can switch from PHP to any other technology (let's say anything from Elixir and Closure to Go and Rust). I assume some developers stay with PHP for serious reasons.
|
|
|
|
|
You missed the point.
Jeremy Falcon
|
|
|
|
|
Yes, I am stupid and I know it 😂
But I know lot of people that think just because something is mature, it must be worthless or dead. By the way, I think Fortran really just got resurrected from it's death.
modified 15-Mar-24 12:06pm.
|
|
|
|
|
Well, PHP is not Javascript, and that alone saves millions of developer hours and hundreds of polar bears per year.
|
|
|
|
|
Yeah, no. The "omg I don't like JavaScript" joke has been played 1,000,000 times. And it's usually by people who know nothing about it.
Jeremy Falcon
|
|
|
|
|
Oh, I know more about it than ever wanted. As a TestComplete test creator. Which says enough about how widespread the cancer is.
|
|
|
|
|
Ok dude... whatever.
Jeremy Falcon
|
|
|
|
|
Jeremy Falcon wrote: But, it had its day that's over now. Sorry, but the facts indicate that you're wrong.
Reading different web statistics pages, there are over 33 million PHP web sites in existence. Another states that 21% of the top 10,000 sites (however the top sites are defined) are PHP. Please note that the exact numbers don't matter -- they simply indicate there's a LOT of PHP on the net.
Why isn't PHP going away?
* It has critical mass, possibly more than any other language.
* It is fully supported and updated.
* It works, and it works well.
* Newer languages are not better, and none approach the critical mass of PHP.
* The cost of rewriting millions of sites in a different language is not realistic.
Like many others on this forum, I've been in IT for decades, and during that time a lot of languages have come-n-gone. For many reasons PHP is a survivor and will continue to thrive. If I was a serious PHP developer (I dabble in it) I'd not worry about finding a job.
|
|
|
|
|
I've been in IT for over 30 years, and I can tell you firsthand you're not only off base and know nothing about me. But...
- I no longer have the desire to spend time on these types of chats. This is the reason I avoid places like CP way more than I used to.
- You're presumptuous.
- It's a dying language. You wanna squabble over the minutiae fine, but look where the money goes. If you don't understand this concept, I'm not spending time talking about it.
- I said for new projects.
- Critical mass is the only reason it's still around. Again I said for new projects. Not sure why you didn't pick up on that. Guess you're getting emotional. This is the part where you deny it.
- Nothing you can say I haven't heard before.
Jeremy Falcon
|
|
|
|
|
I didn't read the replies first, but will post my response to the question.
I started working on a customers PHP project in Sept 2020, which was written in PHP 4.2 between 2004 to 2010 by some Chinese students from China. They did a terrible job, and everything you can possibly do wrong they did wrong. I can't even read the code. Anyways, it was so bad, that I told the customer that they really need to sort of start from scratch again.
Why I chose pure PHP 7.4 for them. Drum roll please ....
Because of time and money ...
So many of the new technologies are constantly being updated so fast, that it's hard to keep up with the changes. PHP 7.4 is still that Swiss Army knife that can do almost anything, and PHP doesn't bug you to update code, or doesn't require a team of skilled software engineers to maintain over decades. Once you write it and deploy, it can run for a long time without more time and money. I can move on to the next project, and the customer can use their new app for years to come, and pass that app to their next generation of children that will take over the business when they come home from College or University.
Our economy in the US has a strong service sector, comprised of small business owners with family involved, and have become generational now spanning their services across decades of time. I look at pure PHP 7.4+ from an economic point of view, and not from a superior technology point of view, considering time, labor, capital investment and maintenance being more stable and predictable across good times and bad.
I don't agree with PHP Laravel as being a framework built on PHP, but opted for pure PHP, in which I was able to build a near MVC type of Web Application using modern objects and smarter HTML views with the objects, and include robust HTTP request and response handling, and robust security features.
I have no regrets in my decision for my customers path forward, and I'm very happy with the results. Well that's my argument for PHP. May not be what you were expecting to hear or read, but I hope it satisfies your thirst to know why.
If it ain't broke don't fix it
Discover my world at jkirkerx.com
|
|
|
|
|
I'm learning the USB wire protocol. I don't want to. I have to implement a USB endpoint on some hardware and I've never done it from scratch before. What a PITA.
The protocol was clearly designed by committee.
Check out my IoT graphics library here:
https://honeythecodewitch.com/gfx
And my IoT UI/User Experience library here:
https://honeythecodewitch.com/uix
|
|
|
|
|
Just be thankful you haven't got FireWire to do, My first Boss was involved in the definition of the standard.
FireWire I'm sure was a way not define a standard.
modified 14-Mar-24 16:21pm.
|
|
|
|
|
|
I used the Serial Port Complete & Parallel Port Complete for early stuff (VB6!) got a copy of USB Complete and never really used it, all of the USB ports I had to interface to appeared as COM ports.
|
|
|
|
|
I wish you the best of luck. May your code work on the first few tries.
__________________
Lord, grant me the serenity to accept that there are some things I just can’t keep up with, the determination to keep up with the things I must keep up with, and the wisdom to find a good RSS feed from someone who keeps up with what I’d like to, but just don’t have the damn bandwidth to handle right now.
© 2009, Rex Hammock
|
|
|
|
|
My view is, if my code works for the Demo and the acceptance trials it becomes someone elses problem!
|
|
|
|
|
Yes, and I usually was the programmer who picked up maintenance. At two companies I worked for, I specialized in it!
__________________
Lord, grant me the serenity to accept that there are some things I just can’t keep up with, the determination to keep up with the things I must keep up with, and the wisdom to find a good RSS feed from someone who keeps up with what I’d like to, but just don’t have the damn bandwidth to handle right now.
© 2009, Rex Hammock
|
|
|
|
|
Working things aren't simple, and there is no way a thing with Universal in the name can be simple anyway.
|
|
|
|
|
Wasn't really hoping for something simple, just sensible. Actually to be honest I was expecting something designed by committee, and that's what I got.
Check out my IoT graphics library here:
https://honeythecodewitch.com/gfx
And my IoT UI/User Experience library here:
https://honeythecodewitch.com/uix
|
|
|
|