I think one major strength of PHP has always been its “get the job done” mentality, and for the web 1.0 and web 2.0 areas it was perfectly suited, even going so far that an analogy “vibe coding language of the 2000s” could be considered on point (source by Günter Grodotzki), but in contrast to his post I am seeking a positive stance on PHPs relevancy.
In 2026, with AI coding spreading like wildfire, if a language or tool stack gets the job done is still extremely important. But today, none of these are an edge for PHP anymore: a rapid feedback cycle, no compile step, var_dump debugging, easy syntax, good documentation, and the near-perfect simplicity and integration of the LAMP stack
I find the pain points PHP was amazing at solving in the 2000s are either almost irrelevant today, or other stacks have caught up or surpassed PHP in individual areas.
So how does PHP stay relevant to get the job done today?
I am thinking about this question a lot. It’s existential for me and my employees, because my business Tideways builds and sells a profiler that works exclusively with PHP.
You can probably call me a PHP superfan, so not only is my business keeping me in this ecosystem, I also want to see it succeed personally, potentially with irrational passion.
Interlude: The selfishness of liking to type out code
Like the The Primagen stated in his latest video “I am alone”, I have come to think that the craft of coding, writing “beautiful code”, is a selfish pursuit.
There are major benefits of writing maintainable and testable code, have well designed abstractions also with AI assisted programming. But me typing it out and enjoying the process doesn't make much economic sense.
With 27 years of intensive coding experience, I find that a lot of the code that I need to type is already fully typed in my head. It just needs to be serialized from brain into actual written bytes.
Yes, you can argue for separating economics from craft. Take carpentry: there's the automated, tool-supported, economically efficient way, and there's the “crafty” way - which is either too expensive for your actual housing needs, or made and sold as art.
It looks to me that there will be a two-class society: Low level fundamental code, which might very well be open source, is still overwhelmingly going to be written manually, because at the foundation “average results” are not going to cut it. On top of that foundation, more and more code will be written by AI.
This new way of writing code can influence how new programmers decide to pick languages for their projects. And the PHP community could work on a more compelling offer here by simplifying running PHP in development environments, and integrating it more easily with the operating side (container deployments, observability, …).
To a newcomer, deploying a single Go or Rust binary looks far easier than packaging and shipping hundreds of PHP files.
Software is about the value it provides its users
So for PHP (and any other language for that matter) the job do be done is increasingly just the value that users of the software get out of it, and providing the shortest path from idea to production-ready product.
This can be achieved by leveraging existing ecosystems, which PHP has plenty of supply: WordPress, NextCloud, Drupal, TYPO3, Magento, Moodle, Shopware, and tons of other special purpose software that has been built over the last 30 years, often with extensive plugin ecosystems. Millions of non-technical people are already used to manage these tools through their backends.
And on top, they can fall back on thousands of extremely stable libraries with long history that the Composer/Packagist ecosystem provides.
Downside of these ecosystems is often that their architecture and code quality has not kept up with modern application development best practices.
One approach for people wanting to keep PHP relevant could be building modern free software in PHP that launches and creates new ecosystems. Target users should be non-tech users. There should be a strong open-source PHP project in every software category: n8n, openclaw, codeberg. Between 2000 and 2010 chances that these projects would have been written in PHP were 95%+.
In true spirit of the meme “PHP the borg of programming languages”, PHP should continue to embed itself in other domains. Projects like NativePHP have the potential to do this.
Free software and non-corporate governance is major benefit
A second major benefit of PHP is its completly open governance model, its truly free software without corporate interests in control. As my friend Sebastian recently wrote: Digital Sovereignty is written in PHP.
Both benefits also apply to Python or Node.Js (less about Ruby, because its dominated by Rails - which has corporate governance).
Integrating with AI systems: PHP and MCP v2
There are serious problems with AI: copyright infringement, energy usage, valuation craziness, massively driving up the costs of hardware for established companies. But whatever your stance on them is, pandora’s box has been opened on using LLMs for coding and its not going to go away. My hope is that local, sovereign AI will take over sooner than later, hopefully countering big techs efforts to monopolize.
As LLMs and agentic workflows become part of how more people work, integrating with agent harnesses through APIs, CLIs and MCP will only get more important.
Before MCP v2 released in June 2026 PHP was in a bad place. The protocol required stateful sessions, and permanent connections to the server, both things which PHP cannot easily do out of the box. With MCP v2, more RESTful, shared nothing paradigm has been established.
My hope is that this will give PHP projects a big push once they seize the opportunity to provide MCPs.
In Summary
This is not a complete picture, subjectively what would make PHP the tool of choice to get things done in the future?
- PHP should be simpler operationally across the whole development lifecycle Ideas here could be:
- A web server shipped with PHP that is in front of FPM
- a way to start a php server that includes execution of long running jobs, and cronjobs, dependency installation via composer.
- out of the box metrics APIs and better log exporting.
- bundling more capabilities that are currently optional (xml, zip, …)
- PHP would hugely benefit from a new wave of modern off-the-shelf software that is focussed on solving important human needs. As a SaaS business operator I am in a conflict of interest in this regard.
- Pound on the Digital Sovereignty argument: PHP should be chosen because it cannot be taken away from you, independent of its for personal, societal, governmental or corporate use cases.
- With MCP v2, more PHP software will be able to integrate easily with agentic workflows. Writing PHP code that integrates with AI could become as easy as writing PHP in 2021.
What do you think?