Two days ago I was watching the election results on various sites, along with many others. Some sites just didn't work in a slightly older browser, and those which did were still consuming a lot more resources than they really needed. It shouldn't require the latest in web technologies and computing hardware to show a simple dynamically updating outline map.
If the service I'm using is hosted on us-east-1, it's using power from virginia, which uses a mix of natural gas and nuclear. [1]
Based on that... is running more logic on my computer or on an edge server within Québec, actually using less GHG-emitting energy than running it on the origin server?
[0] https://www.cer-rec.gc.ca/en/data-analysis/energy-markets/pr...
[1] https://en.wikipedia.org/wiki/List_of_power_stations_in_Virg...
I think the tools we have nowadays are perfectly fine. It's a matter of how they are used. And I am pretty sure efficiency is not what companies think of when they launch a product.
You can write Java without using incredibly huge class, function, and variable names. You don't have to apply complex design patterns everywhere, or use a complicated framework you barely understand. You don't have to write code that produces 50-line stack traces. I don't have to import a dozen dependencies each with a dozen dependencies of their own, creating an endless security update treadmill. You don't have to write code that needs a gigabyte of RAM for the smallest microservice. Problems like long garbage collection pauses can be solved.
But if I take a job writing Java software? Probably I'm going to be handed a codebase written the way most Java developers write Java code. And I'd better make peace with that, or find a different job.
I think this is not how companies work though, because they do not operate based on your views alone. And getting all people to agree on some of the topics you mentioned (e.g. importing dependencies vs. rolling your own implementations) is an incredibly complex task.
Yes, it doesn't have to be that time frame, but it would be a good reference point. Early 2000s allowed us to do most of what we can now, there are certainly exceptions, but the average website would be no worse. The savings in processing and memory consumption can then either be used to run other things, or extend the usable lifetime of a device. There's no reason why we could not use the same device for 10 or more years, again with some specialised exceptions.
Give that this is specifically a w3.org SIG, I'd suggest doing a LTS web standard, something like 10 - 15 years. Make it have a reduced feature set in terms of Javascript and CSS. For some businesses it would be attractive to know that a solution developed to a specific standard which would mean compatibility across devices and software for 10 years (ideally more, 10 years isn't that long). Newer devices would consume less power and older devices would require less frequent replacement.
The problem is that this would need to find it's way into a browser, which would also need a long term supported and stable operating system, to gain all the benefits.
As to HTML + CSS: it goes other way around: HTML and CSS (or rather its subsets) are being integrated into "normal languages", like Qt, Java. I'm not sure I ever saw any technology that could serve as a replacement for HTML + CSS. May be Eclipse RAP or Blazor? But they are so heavy that React will look like a butterfly and they're not aiming to replace HTML/CSS but rather just use it as output medium for their UI.
Yeah that kind of stuff is closer but most of those types of frameworks seem to lean heavily on mapping to JS because they try to be turing complete rather than just being easy to work with DSLs.
- Browsers don't "talk" any other language but JS
- All browser APIs are exposed through JS only
- You can't manipulate DOM except through JS
- You can't do "fancy web flourish" without manipulating DOM. If you target Canvas/WebGL/WebGPU, you'd have to first create your entire graphics lib + flourish + font handling and rendering + accessibility + ... from scratch. And load all that on every page load
- Any language compiling into WASM would still need JS-integration for any of the above. Including Canvas/WebGL/WebGPU
- Any language without JS and DOM semantics will need to account for that (e.g. GC on DOM nodes)
- Any language compiling to WASM would need to load its runtime to actually run (including any libs). See network tab for any such project. e.g. Blazor
I'm not looking for browsers to talk with any language. I'm looking for some DSL that directly maps to the 3 layout languages that browsers understand (i.e. HTML + CSS + SVG). Not anything turing complete at runtime but rather a sane way of describing a webpage layout with fancy styling, UI elements, transitions, and animations but without dealing with the pain that comes with actually writing in the native browser layout languages.
> You can't do "fancy web flourish" without manipulating DOM.
There is a lot of web flourish you can do without manipulating the DOM. It's not actually terribly unperformant to do but writing that code (mostly HTML + CSS, occasionally SVG) feels like peeling your eyelids with an unwashed lemon zester.
> ...
And for the rest of that, again I'm not looking for anything that actually executes in the browser. Just a sane, modern layout language that compiles down to static HTML and CSS with no JS or WASM (unless you explicitly ask for it).
Just some domain specific language for writing sane low-to-no-js web pages or parts of web pages without having to manually fiddle with HTML or CSS in any real amount.
Static site generators honestly get me a lot of the way there but those are unfortunately template based which means any significant customisation requires dealing with the HTML and CSS rather than being able to just describe the layout and behavior I want.
Honestly I'd just write the compiler/lang myself if I didn't hate frontend so much.
Depends on what you mean by sane modern layout :)
Many modern layouts are still impossible without a lot of JS intervention. Many web flourishes also require Javascript :)
That's why there are no DSLs for this: HTML and CSS already are the DSLs you're looking for.
My gut feeling is that you cannot (fully) abstract away HTML/CSS if you want the result to feel like an actual website.
With Astro, MkDocs or docfx, I do not have to touch HTML, except maybe for creating the master layout and/or transformation rules, if needed.
Because you can't use markdown to design CSS or SVG animations. I want to be able to design an animation with code that describes how the elements move/interact and then compile that down into CSS or SVG keyframes so that I don't have to manually declare however many arbitrarily complex keyframes.
I understand the purpose of a static site generator and I'm not looking for anything to replace that. Rather I want to be able to write the templates for my static sites without 1. writing exceptionally tedious HTML & CSS and without 2. relying on client side JS to do those tedious things.
JS crypto miners working overtime. Will allow CNN/MSNBC/Fox/etc. to stay in business for another 4 years.
(Along with Ozempic ads, which are basically completely unnecessary now.)
HTMX. Or basically we need some JS framework like HTMX / Alpine.js where we could test what to include in HTML.
"Our new update means we reach WGAC 2 Level AA to > 90% and WSG to 60%, next release we aim to reach WSG to >70%" might be something we hear next year.
I completely agree. However, I think browsers are also to blame in some part.
On web sites that I build, I sometimes get alerts from Safari that my page is bogging down the computer and it offers to "reduce protections" to make the page perform better. But this is always on pages that are plain HTML and CSS, and don't even have animations. No Javascript. No canvas. Not even forms. And the total payload is often less than 20K.
I don't know what else I can do to make it lighter.
Well...to your intended purpose. They're often better aligned with the purpose of keeping the business running.
Maybe a badge/score to say how well a site is for efficiency, at least for the front end. Unused code etc being deductible from say a score of 100.
I'm still running the original iPhone SE from 2016, and there are basically two things that will reliably heat up the phone and absolutely destroy the battery: news websites, and the github web frontend.
It's pathetic how many resources these things use when their main job is to essentially display some text to you. The github native app works completely fine which shows it's not a problem with the phone, it's a problem with devs not caring at all about performance.
So it’s not only the resources needed by page, but that older devices end up in landfills.
How will w3c sustainability group run by people from irrelevant organizations help with that?
Google is responsible for one of the largest chunks of bloat with its ads, embeds, tag manager, analytics etc. And they couldn't care less. They could penalise sites, but instead they now say that loading a page in under 2.5 seconds is fast: https://blog.chromium.org/2020/05/the-science-behind-web-vit...
When you live next to a Google or Facebook data center some reality starts to set in. They easily consume most of the output of a single small urban power plant on their own. It’s nuts. I didn’t realize how nuts it is until someone explained it to me, in my part time job I work with a Houston based power company lawyer that specializes in contracts for data centers. I doubt those massive data centers are reliant on JavaScript.
As for JavaScript there is a simple solution that works wonders in every other industry: licensing and liability. The code is bad because the developers that write it are shit. That’s never going to change until businesses have a financial incentive to train for competence. All the wishful thinking about less JavaScript is just more virtue signaling.
So, measuring their power consumption isn't going to indicate anything unless you know what exactly does that datacenter support.
At one point I ran across the statistic that said that Facebook's internal network had more IP addresses than the entire public Internet. Maybe whoever claimed that exaggerated, but all the Web stuff Facebook has to offer doesn't need even a single percent of IP addresses of the entire Internet.
Also, as far as I know, Facebook has its own h/w ambitions. It builds its own h/w and equips its own datacenters with it. While this is definitely not their main business, h/w development is a lot more resource-intensive than s/w development when it comes to datcenter usage (when it's about developing datacenter h/w).
If the datacenter in question isn't involved in Web at all, why would it be of any interest how much power it consumes?
“Since the advent of the modern web, the ability to include embedded fonts and provide a more customized experience has seen their use explode. They aren't always the most performant option (which poses emissions hazards) and come with a few issues such as Flash Of Unstyled Content (FOUC) / Flash Of Unstyled Text (FOUT) which should be addressed.”
IMO, if we want reduced emissions, citizens in all countries need to tell our leaders/representatives that the monetary cost of polluting must increase, until emissions are drastically decreased – i.e. we must internalize these negative externalities. In the EU, we have the Emissions Trading System for this purpose.
If we don’t demand this from our leaders, how can we expect emissions to decrease?
I’m sure a group like the Sustainable Web Interest Group can come up with a bunch of nice ideas, but I’m not convinced they can solve climate change.
Sure, stuff like embedded fonts might possibly increase emissions. But if W3C are advicing against their usage, where’s the data that supports this guideline?
(A Pigouvian tax can be another alternative, but harder to implement in EU, since taxes here are collected on the nation-level.)
This is usually the point where whataboutism strikes and people "require" conversations around what constitutes a popular font. Browsers are already full of analytics and can record this. Google Fonts serves probably billions of font requests a day, so they can record this.
Have the usual Big Tech bunch agree to start shipping the top, say 100, most popular fonts in their OS and/or browsers.
If it was possible to updated old devices with any custom OS, many devices would continue to work instead of being disposed of.
I wish I could jail break it or install an alternative os. It’s only useful to play the few games I have for it (most of those haven’t been upgraded to 64 bit so don’t run on new OS and hardware). It’s good hardware but I feel very much at the whim of Apple.
— “The leanternet principles” <https://leanternet.com/>
— “The 250KB Club - The Web Is Doom” <https://250kb.club/>
— “Sustainable Web Design” <https://sustainablewebdesign.org/>
— ”Other 'clubs'”: no-js.club, 1mb.club, 512kb.club, 250kb.club, 10kbclub.com, 1kb.club, js1k.com, js1024.fun
— “CSS Minecraft – written in 100% HTML/CSS with 0 javascript” <https://benjaminaster.com/css-minecraft/>
— “Low Tech – a solar-powered website” <https://solar.lowtechmagazine.com/>
Since Manifest v3 shame there's no chance web would get any less ads. Only more. Much more.
Was I the only one thrown momentarily by the use of "web worker" to refer to a human?
Just get rid of all the ads and related code and infrastructure. and all the extra calls here and there for tracking and spying.
That would save a shitload of required processing and network traffic.
Internalize the costs of energy as a first step. If you manage to make web fonts cost $2 per load, people will find their own ways to use less of them. If you make web fonts CHEAPER to load by making them "more efficient," then people will use MORE of them!
I also had to do a X icon to replace the Twitter bird. So I went to get the official one and make it into my lean SVG. Again, you would not believe how much bloat was in what should have been a very simple file.
This is no rant about Twitter, the web in general is 99% bloat. I don't believe Google have 'stewarded' the web well enough to keep it lean.
If we go with the icon example, an icon has to be simple or else it is not an icon. Yet we have huge icon sets as fonts with excessive bloat. This is why I end up having to hand-carve SVG assets on the regular.
This aspect of simplicity applies to web pages too. Style sheets should not be thousands of lines. Content does not need to be nested in a billion divs, particularly since no div elements are needed now we have content sectioning elements and CSS grid layout.
The leanness of a website should be important as an expression of brand values for companies. For example, if your business is making cars, your website should be the fastest loading one to reflect your 0-60 times.
Hopefully we will get metrics for efficiency as one of things like accessibility that people strive for in varying degrees, with this efficiency being good for SEO. As it is, Google prefer data to be poorly structured as wading through rubbish is what their business depends on. If all content was well organised without the bloat then others would be able to do search to compete with Google. Hence we have a sea of divs on every web page, even though MDN docs says the div element is the element of last resort.
ffmpeg -i 'https://video.twimg.com/ext_tw_video/1846357395959615488/pu/pl/ecNx-sTzYA9doHYO.m3u8' -analyzeduration 5G -codec:a libmp3lame -b:a 96k output.mp4
(Actually I just checked and it also supports downloading Twitter videos directly.)
Given current trends, that seems unlikely. If anything, with LLMs that can do both CGI and storycrafting, it will get even less efficient as content is generated rather than serving stored files.
Still, this is important too.
I personally see different problems in many of the areas, but I’d like to know if someone has already organised these topics.
A nuclear and solar grid powered 3 tons vehicle isn't much more sustainable than an ICE one if it's still carrying 1.1 human on average.
I'd love to know the net effect of all of that. And the "digital industry" in the first paragraph: is that everything digital globally, including crypto? Or is it just the stuff this WIG can address, i.e. the Web?
This was nearly entirely bureaucratic bollocks, but here and there you can parse some useful information. I think.
Was this written with the aid of AI? It seems having an AI that summerize all of it would be a big win.
I will nominate AI for the role of 'Big Oil'.
Exposing the Sustainable Development Goals
We are halfway through a plot to seize control of the world. It may seem like a conspiracy theory, but it's placed awfully prominently, and everywhere, to be such a thing. The United Nations Agenda 2030 is a sweeping program to take control of our entire world. It launched in 2015 with an ambitious "17 Goals to Transform our World" and 169 targets to hit by the year 2030. That was eight years ago, and we can get a sense of how it's going. Badly. Tyrannically. Farcically. Reading from the Agenda announcement itself, in this episode of the New Discourses Podcast, host James Lindsay introduces the 17 Sustainable Development Goals of United Nations Agenda 2030 and shows how every one of them grants the pretext to seize control over the world and all human life and activity in it. Join him to know your enemy.
I.e. blaming the Internet for being one of the greatest polluters seems disingenuous, because... what if 90% of that pollution comes from Web? So, maybe the Internet works fine, but the Web needs fixing?
""the IG plans to liaise with regulatory bodies to improve compliance targets""
Regulatory bodies absolutely do not care about W3C. Hell, they barely care about the IETF, IEEE, ICANN, etc.
I'm all for pushing for sustainability, but look at the other interest groups. For example, privacy. Cloudflare just published an article talking about post-quantum crypto [1] where they talk about how wild a percent of traffic would be just cert exchange (and, currently already is). There will always be competing interests, so a body that only exists to checks notes talk about ""sustainability"" on the web feels moot.
They explicitly say hardware is out of scope. Cool. So software. The only way to help sustainability is to use less or make it more efficient. Less never happens, and efficiency isn't a concern above ad revenue for literally anyone.
Honestly, I'm inclined to see this as actively harmful more than anything. Putting out statements about sustainability just dilutes the waters on web issues they might have real pull in, like standards for user privacy that DO help with sustainability. For example, making it easier to choose what content gets delivered cough DNS blackhole adblock cough means less data being transfered.
I still wish this group the best and hope that they can discuss actions of other groups (Such as the Media and Entertainment Interest Group) in context of their choice of standards impact on processing power requirements.
Honestly, reading the manifesto [2] just makes me more angry. It doesn't say anything. Go read some solar-punk manifestos by people on the Indie Web or in Solarpunk culture. Those at least say something. This is just marketing fluff for the sponsors at the bottom of the page.
[1] https://blog.cloudflare.com/another-look-at-pq-signatures/ [2] https://www.sustainablewebmanifesto.com
I suspect it will come as news to you that many governments do base laws and regulations on W3C https://www.w3.org/WAI/policies/ including EU and US Department of Justice https://www.ada.gov/resources/2024-03-08-web-rule/
I keep thinking they would do better if they got ahead of things and suggested a toolchain for future projects, that would increase the odds that they get adopted.
Getting a few groups of volunteers together to learn a handful of LTS technology stacks instead of a cartesian product of all of them that grabbed two people's fancy three years ago and now they're bored/out of money. It would make it a lot easier to get to a more PBS-adjacent model of internet for the public good.
In some respects this is a different sort of sustainable than what they mention in the article, but amortizing a bunch of relatively low-pop services across a single cluster and admin team still counts as an efficiency, versus having them scattered on disparate hardware, disappear from neglect, to be recreated again in a few years from scratch, after someone squats the old URL and refuses to give it back.
They published a charter. They're going to establish guidelines for sustainable web development and tools for measuring your impact. Yes, static architectures will probably be one path for improvement.
> There will always be competing interests, so a body that only exists to checks notes talk about ""sustainability"" on the web feels moot.
I'm not following this point. The existence of entrenched interests means that no opposing interests should be researched? Why is "sustainability" in quotes, is it not a legitimate pursuit, or are you implying that they have ulterior motives?
> They explicitly say hardware is out of scope. Cool.
Hardware is out of scope "unless related to hosting & infrastructure," AKA the cloud. That is an absolutely massive scope within the hardware realm.
> Honestly, reading the manifesto [2] just makes me more angry. It doesn't say anything.
It sounds like you're looking for the guidelines that this group aims to publish. A manifesto in this context is not intended to be a solution or a prescription; it's a framework for alignment towards a goal. The concrete solutions are the goal of the group.
https://newdiscourses.com/2021/10/sustainability-tyranny-21s...
Sustainability is going to be the buzzword of the century. Everywhere we turn, we hear about sustainable practices in business and industry, sustainable foods and agriculture, sustainable energy, and so on. Businesses and governments sign on to “Sustainable Development Goals,” and so civil responsibility is framed in terms of this seemingly simple idea: sustainability. What does sustainability entail, though? What informs it? In this episode of the New Discourses Podcast, James Lindsay walks through Herbert Marcuse’s New Leftism of the 1960s and 1970s and explains how sustainability has become Marcuse’s “New Sensibility.” In other words, sustainability is the new way of thinking about the world so that we can have liberation, which is to say Communism. Join James in this groundbreaking episode of the New Discourses Podcast to explore this idea at its ominous roots.
Yes, sustainability will mean different things across different domains, because for some reason people are starting to realize the consequences of their actions in the real world. What else would you expect?
> The products and services we provide will use the least amount of energy and material resources possible.
Is this from the same W3C that has been pushing us all since 2013 to upload our locally hosted files to one of 3 major cloud providers who just happen to be megadonors to W3C? Funny now that we have to send our personal files across the internet. I wonder what the sustainability "under/over" is gonna be when I have to send packets around the world to retrieve the files that used to live on my computer.
https://www.w3.org/WAI/RD/wiki/Cloud_Computing_Accessibility...
Oh, and it starts with a giant disclaimer that says "This Wiki page is edited by participants of the RDWG. It does not necessarily represent consensus and it may have incorrect information or information that is not supported by other Working Group participants, WAI, or W3C. It may also have some very useful information."
Do you have anything else to point to to suggest that the W3C is "pushing us all since 2013" towards 3 cloud providers?
There is no alterntive between storing everything in the cloud and smoke towers.
(still, I assume not the cloud storage is the most energy intensive thingy out there - but perhaps the processing of those for whatever agenda, and else - but the w3 signals are mixed the least. Perhaps this is from some sort of common corporate script book distributed in the MBI courses, from the chapter "how to pretend being serious environmentalist", mixed with the other one "deflect inconvenient/expensive steps into the infinite future or never by forming an interest group")