Every generation of programmers has its moment of declaring PHP dead. It happened when Ruby on Rails became the craze back around 2008, it happened again with the Node.js boom, it happened once more when Go showed up promising compiled-language performance with scripting-language syntax. And here we are, almost in 2027, and PHP is still there, quietly running the largest slice of the internet any language has ever held on the server side. So the question worth asking isn't "did PHP die?" — it's why we keep asking that every couple of years, even when the numbers keep saying otherwise. Let's start with the numbers, since that's usually where the conversation stalls. As of July 2026, W3Techs had PHP at 70.6% of sites with a known server-side language. More recent readings of the same survey push that up to somewhere between 71% and nearly 80%, depending on how each source slices the sample (whether or not it includes sites with no detectable CMS, for instance). Second place in that race is Ruby, at 6.7%. Server-side JavaScript comes in at 5.9%, Java at 5.4%, ASP.NET at 4.4%. That's not a narrow lead. It's territory no other language comes close to contesting. And this isn't recent or the result of some passing spike — PHP's share has hovered in that range for years, surviving hype cycle after hype cycle without really losing meaningful ground. Here's where the more interesting part of the story lives, and it's the reason so many people remain convinced PHP is a thing of the past: popularity in developer forums and actual usage in production are two completely different metrics, and the "PHP is dead" argument has almost always leaned on the first one while ignoring the second entirely. You open Twitter, Reddit, Hacker News, and you see a flood of people excited about Rust, about Bun, about whatever framework of the moment promises to reinvent the backend. That's the conversation. That's the hype. But hype isn't the same thing as what's actually running when someone hits F5 on some random site out there. A big chunk of that dominance has a name: WordPress. Over 43% of all websites on the internet run WordPress, and WordPress is PHP down to the roots. That means PHP's massive share isn't coming from startups choosing the language today, in 2026, for a brand-new project — it's coming from an absurdly large legacy ecosystem: personal blogs, small-business institutional sites, WooCommerce stores, regional news portals. Stuff that's already live, already working, that nobody has any real reason to rewrite from scratch just because a trendier language showed up. Then there's another point that tends to slip past anyone looking in from the outside: the job market doesn't treat PHP like a dying skill. UK job data shows that generic PHP roles have softened a bit recently, that's true, but senior and Laravel-focused positions are still paying well — and that's not the behavior of an ecosystem being abandoned, it's the behavior of an ecosystem specializing. There's also the technical argument, which anyone who hasn't touched PHP since 2015 tends to ignore completely. Today's PHP isn't the spaghetti PHP we all remember from some college project. Starting with version 8, the language picked up JIT compilation, union types, attributes, named arguments, match expressions — things that put modern PHP in a pretty different league from that loose script tangled up with HTML that defined the language's reputation for so long. Anyone who still sees PHP as a synonym for bad code is usually judging the old reputation, not the current state of things. Now, saying "PHP didn't die" isn't the same as saying nothing is changing — and this is where the conversation genuinely splits, without a single closed answer on either side. A meaningful chunk of the market is shifting APIs and real-time features over to other runtimes, while still keeping the content layer, the CMS, the "surface" part of the site running on PHP. In other words: PHP isn't being fully replaced anywhere — it's being pushed into a more specific role, more of a content backbone, while other stacks take over the parts that demand heavier processing or real-time communication. Is that decline? Or is it just a division of labor, each tool doing what it does best? Depends who you ask, honestly, and I think that's the most honest answer there is to give. One thing worth keeping in mind before repeating any argument in this debate: be skeptical of anyone talking about PHP's death without bringing a single number along. It's basically a pattern — the claim shows up as a loose opinion, repeated because it's been repeated before, not because anyone actually checked whether the data had changed. And on the other side, be just as skeptical of "market share" stats thrown around without context, because market share across existing sites and adoption in brand-new developer projects are pretty different things, and each side of the debate tends to reach for whichever metric already backs up what it wanted to prove going in. At the end of the day, the numbers don't leave much room for debate: PHP didn't die. It just lost the shine of hype, which is a pretty different thing from losing relevance. And if you work in development, or you're trying to decide whether it's worth learning or keeping PHP on your résumé, that distinction between hype and actual usage might be the single most useful thing in everything you just read. So now I want to know from you: would you actually pick PHP today for a brand-new project from scratch, or are you just still using it because it's already there and working? Drop a comment below — this discussion always gets people talking.