We're approaching the point where speed is essentially a solved problem unless you're at Google-scale.
The blog post https://www.techempower.com/blog/ puts it better than I:
> I argue that if you raise the framework's performance ceiling, application developers get the headroom—which is a type of luxury—to develop their application more freely (rapidly, brute-force, carefully, carelessly, or somewhere in between). In large part, they can defer the mental burden of worrying about performance, and in some cases can defer that concern forever. Developers on slower platforms often have so thoroughly internalized the limitations of their platform that they don't even recognize the resulting pathologies: Slow platforms yield premature architectural complexity as the weapons of “high-scale” such as message queues, caches, job queues, worker clusters, and beyond are introduced at load levels that simply should not warrant the complexity.
That said I disagree a bit with saying it is a a solved problem unless you are Google-scale. In developing countries if you are in a small company without big investors, you need to do quite a bit more juggling with infrastructure budgets once you are past MVP compared to for example where I'm from Switzerland.
But are people really doing python and web development?
I love python for my web crawlers, photoshop automation, and table formatting automation- But I never considered it a web contender.
Any major sites running python? Whats the popular way people are doing it?
YouTube is quite a popular one.
For fast Web App development it's a superb language choice.
Not so good if you wanted to write a super-high-throughput proxy service, but asyncio & ASGI is starting to change things there too.
> Any major sites running python?
Instagram is probably the biggest headliner these days, but there's stacks of others.
That's in my opinion the main reason why Python and Ruby on Rails got massively popular for web dev in the 2010s. Since then new and existing languages have started breaching into that spot.
What still makes Python stand out is its flexibility and the sheer amount of resources available(libraries, frameworks, books, etc.) across numerous domains. It is a good enough choice for a lot of things, although its not necessarily the best choice for anything.
Jokes aside, I do agree that from a developer's standpoint, the optimal way to act (optimising for business success) is as you say, ignore performance at first. But from an engineering/craftsmanship standpoint, it should really bother you to have an inefficient system.
Just think of all that extra CPU load, and all the CO2, all the coal burnt to sustain it ...
366 rps in a test scenario doesn't scratch the surface of what I need, and I'm not doing anything I'd consider crazy scale. When you need to support hundreds of thousands of rps, small improvements can be very noticeable.
Becoming a Fortune 1000 company isn't "most apps"
When I'm a fortune 10,000 company, we can afford to rewrite the web app.
You have no idea what you're talking about. Not only did they produce TWO different implementations of PHP (HHVM and Hack), a vast majority of their infrastructure is NOT in PHP (which is mostly reserved for the front-end/web tier).