Follow

The web browser situation isn't sustainable for free software. the web standards committee needs to dial down the complexity of the specs so that they can be independently implemented

@Shamar @rain What needs to be done is that a new standards committee be formed out of an alliance of web practitioners. For example, invite anyone developing a fediverse server to join and give them edit permissions on a wiki.

@rain agreed but good luck, considering the w3c has already pushed for a standard DRM API. Time to question authority! 😅

@mxjxn @rain

Good point. Where do we go if the w3c becomes a puppet.

@rain I'd say the problem is with the quantity, each individual isn't bad at all.

@charlag @rain I won't argue you there.

But you know what I want: replace JavaScript (and ideally cookies) with something resembling Intercooler.js (intercoolerjs.org/ ). Seriously check it out!

I'm not sure if that's enough though.

@alcinnz @rain that's pretty cool.
I want both: simple document model and widely supported app platform and until we split it web is going to be torn apart in two directions.
There's also tons of complexity for things no one really asked for like Web Components or SharedArrayBuffers.

@charlag @rain Certainly! And once we do that both can be a lot simpler.

From my understanding of computation, all that's really needed for an app platform would be a bytecode language and a window manager/bytecode library. Everything else can be implemented inside the sandbox.

@charlag @rain That second "bytecode" was meant to say "graphics".

@alcinnz @rain truly.
there are several things that I, as an app developer would like to see in app platform:
- Any efficient bytecode interpreter
- UI toolkit which doesn't suck. People say it's a myth and cannot exist but I'm hopeful. I shouldn't have to roll out my own RTL supprot e.g. or accessibility toolkit
- Low-level storage which doesn't suck. People say... well, you know
- Permission model
- Communicating with other apps

@charlag @rain That sounds great!

Personally though I'd implement the UI library inside the sandbox upon more basic primitives, but that effort should be shared between apps.

@alcinnz @rain I have "for"s for both low-level and high level .Like if you use default list components in Android you get Accessibility scrolling for free. Similar with other things like spanned text.
On the other hand I've seen iOS trying to bring too high-level components, like predefined list cells or heavily restricted toolbar buttons and that just goes in your way.

@charlag @alcinnz @rain Web Components is an attempt to eliminate the complexity of the platform by moving new development from the standard into libraries — so things can be "implemented inside the sandbox" as @alcinnz was saying.

@kragen @alcinnz @rain to me it seems more like "oh, they're using these expensive frameworks. That's because they like components!" while components is being only means and not the end goal. Web Components sound like a horror to me with each black box bringing its own dependencies.

@charlag @alcinnz @rain Alex Russell was trying to do Web Components since NetWindows in 2000 or so, which eventually became Dojo following a C&D from Microsoft. You may be right about *other people's* motivation for *accepting* his proposals at long last.

@rain You can't dial down the complexity of specs; deployed platforms change in only one direction

@kragen @rain i fav'd both of these points, and will add:

this is very much in feature-not-a-bug territory for the browser vendors that own the space (namely: google) and most of the major cloud software providers, so no one with any power has incentives to behave differently.

Sign in to participate in the conversation
niu.moe

Welcome to your niu world ! We are a cute and loving international community O(≧▽≦)O !