That's incorrect. Different sites have different needs and the devs can't pre-program which of the billion sites need what.

Believe it or not websites don't work the same way they did "in the old days". But yes, a lot of it is not needed for many websites. Many of them do require it to function properly though and if devs don't make it available no one will use them.

Just yesterday I was on a news website. I wanted to support it and the author of the piece so I opened a clean session of firefox. No extensions or blocking of any kind.

The "initial" payload (i.e. after I lost patience approximately 30s after initial page load and decided to call a number) was 14.79MB transferred. But the traffic never stopped. In the network view you could see the browser continually running ad auctions and about every 15s the ads on the page would cycle. The combination of auctions and ads on my screen kept that tab fully occupied at 25-40% of my CPU. Firefox self-reported the tab as taking over 400MB of RAM.

This was so egregious that I had to run one simple test. I set my DNS on my desktop to my PiHole and re-ran my experiment.

Initial payload went from almost 14.79 -> 4.00MB (much of which was fonts and oversized images to preview other articles). And the page took 1/4 the RAM and almost no CPU anymore.

Modern web is dogshit.

This was the website in question. https://www.thenation.com/article/politics/welcomefest-dispatch-centrism-abundance/

Yes, but the manager with a shitty MBA doesn't care about overall company appearance of performance, as long as their department looks good on paper. And they figured that would be easier by using four different external libraries, and then let another department figure out the rest.

One of the biggest reasons websites need to run JS is submitting form data to a server. Like this website.

No. Forms function quite perfectly without JS thanks to action=.

Now whether you want to get "desktop app" fancy with forms and pretend you are a "first-class desktop citizen" that's a skill issue. But submitting form data, by itself, has not required JS since at least 1979. Maybe earlier.

Another major reason for using JS is dropdown menus and panels. You need to either write code to listen for the click and reveal/hide it as needed, or you have to do weird CSS tricks that are usually inferior in UX to a JavaScript implementation, or you have to bastardize the form dropdown selector into your general purpose dropdown.

Look for text "HTML's got expandable sections baked in"

I had actually given up on expandable sections for my website (because I didn't want anymore JS than the dark/light switcher I made) until I found this.

Let's be careful how we phrase things here. JavaScript form submission and navigation are choices, not needs.

Also, progressive enhancement / graceful degradation exists. When competent developers (or bosses) want script effects on our sites, we can include them and make the sites continue to function with scripts disabled. It might require more work, but it is absolutely possible.

Framing the script-based approaches to these things as if they were needs contributes to the problem, IMHO.

(I am referring to the vast majority of web sites, of course, not special-purpose web applications like games.)

I miss websites with simple text, links and graphics. We could navigate them perfectly fine without any JS, any dropdowns or whatever. They just displayed the information you came for, nothing extra.

I agree with you about dropdown menus being something that could/should be natively available to HTML, but I'm less convinced about form submission. Sure, if we assume everything is happy path it's a great idea, but a system needs to be robust enough to handle a variety of cases. Maybe you want to redirect a user to a log-on page if they get back a 401, or present an explanation if they get a 403. A 5XX should usually display some sort of error message to the user. A 201 probably needs to add an element into the page, while a 200 might do nothing, or might alter something on the page.

With the huge range of possible paths and desired effects, it pretty quickly becomes apparent that designing an HTML & CSS–only spec that can meet the needs is infeasible. There's definitely a case to be made that JavaScript has become too powerful and can do too many potentially dangerous or privacy-invading things. And maybe a new range of permissions could be considered to limit a lot of that at a more fundamental level. But what we're talking about here with the form submission stuff is the real bare-bones basic stuff JavaScript was designed to make easier—alter the contents of web pages on the fly in response to user actions. And it's really, really good at that.

The web app could switch to lower res images (etc) if your connection is weak. Or if your battery is low it might switch out YouTube embeds for clickable images instead.

The one I use is part of a hardware UTM, but I also use Lockdown VPN on iOS, and https://pi-hole.net/ in a container on my LAN, and then VPN all my devices to my home network when I’m not at home.

Depends on the browser/OS.

My go-to for general browsing is Firefox with uBlock Origin and NoScript, which I also use in Edge; I have a few browsers that are still using uMatrix, and I have a proxy filter that strips calls to .js URLs by default except for specifically allowed URLs.

This was the only one I could think of a good reason to track. I immediately thought of all the grandparents and tech illiterate people who'd probably implode if they had to pick .exe vs .deb vs .dmg/.app (I actually had to look up what MacOS uses...) vs etc. And don't even try to have them guess intel vs amd.

Automatically guessing the operating system saves us tech people from having to figure out they downloaded a file for a completely different OS.

Operating system and CPU architecture are useful for sites to serve the correct binaries when a user is downloading an application.

Barely. You could trim down the data to incredibly low granularity ("OS: Windows", "CPU: Intel Desktop") and you'd still get the exact same binary as 99% of the people 99% of the time, anyway.

Oh yes, that's a very good point, actually. That actually seems such a fundamental use case that you could almost justify it being available without a permission.