Ask HN: Getting tired of complexity in web development

Ultimately, that's what all these different frameworks actually are though. Look at how many different languages over the years have compiled down to JS? CoffeeScript, TypeScript, Elm. And all the frameworks that have smoothed over the native DOM API: React, Angular, JQuery, too many to list.

They all implement these concepts in Javascript. But if we don't standardize and allow browser vendors to implement at a lower level, we will always be taking a speed hit and on the framework treadmill. We need to take what we know works and do it.

I tried out PHP recently for a project because I have fond memories of it, but didn't end up going with it. I want server side rendering for the SEO benefits, but I also want some good 'app like' interactivity on the page. Doing the app like stuff with PHP is possible, but you kind of have to have mirror routines, one to load the data, and other api calls to update the data using whatever binding library you choose.

The good client side rendered js frameworks, react/svelte/vue have SSR solutions, but they also have their own servers for enabling SSR - next.js and sveltekit. So if you need SSR, SEO, good client side interactivity - you might as well as used those pre-packaged servers. You also have the added benefit of working in a single language across the stack. Laravel's front end recommendations aren't great - https://laravel.com/docs/9.x/frontend

I do like how react packages html/css/js as components and that you get static typing across the typescript to html bindings which makes the code more maintainable. Also react has a pretty good front end component library with MUI. You kind of have to 'kitchen sink' the front end libraries with PHP as there's no compiler to weed out the ones you're not using - plus you need to use a CSS library, and 'bring your own javascript' to utilize them with where MUI includes everything wrapped as components already.

Interested in others thoughts on this.

Simply this

>Why does this happen?

I highly disagree with everyone else. This is the answer why:

Because there's no clear or definitive answer to What's best the "best" architecture or system for building these things. The entire industry is full of people making up stuff without anyone truly knowing anything.

For the shortest distance between two points we can calculate an answer. there's a definitive optimum. The question of what's the best web service is infinitely more complex. The inputs are huge in number and even the definition of "best" is complex and hard to define... So in short we essentially have no mathematical theory for software design and technology that can help us define what is "best." It's the same problem they have in Art and Design.

So nobody actually knows what is better BUT they think they know so they build wrappers around bad technology or they build completely new frameworks NONE of which actually provably move the needle forward. In fact the needle can move backwards and nobody knows!

When something can't be calculated, it's "Designed." This is the difference between "design" and calculation. Anything that's designed is basically some human wandering around in an area almost aimlessly without ever truly knowing what is optimum. They just use their gut and find some sort of answer that somewhat works. This is the current and past state of the programming industry. That is why on some intuitive level we can see that a lot of the industry looks like it's moving backwards.

Let's say we do have an equation that can take all the components that go into web application development and give you a numerical rating on how good it is. If we had such a model, my best guess is that if we took a system from the past and compared it to a system now, the Modern system will actually have a LOWER numerical rating.

But that's just my guess. As of now, nobody can truly define what that numerical rating represents and nobody knows whether some new framework or design methodology actually increases the amount of points. We are doomed forever to exist in this endless design loop until someone finds a mathematical model that can prove and layout the big picture from a quantitative perspective.

I think it's very easy to not work there, just leave and don't go back. I suspect most people that stick around either don't know any better or do it for the money/status/whatever and not for the technology.