CSS and design systems people - BlogFlock

Sustained attention/revision - Robin Rendle

2026-05-12T17:02:28.000Z

Writing, I told them, isn’t supposed to be easy, and of course it can be tedious but that doesn’t make it rote. Writing isn’t just the production of sentences – it’s the training of endurance by way of sustained attention. It’s a way of learning what one thinks by attempting to say it. An LLM can reproduce the appearance of that activity, but it can’t replace it, because the value lies not only in the object produced but in the transformation that occurs during its making. [..] I want to see what happens when someone tries to move through language without an intermediary finishing the thought.

This reminds me of this piece from Saunders:

This mode of revision, then, is ultimately about imagining that your reader is as humane, bright, witty, experienced and well intentioned as you, and that, to communicate intimately with her, you have to maintain the state, through revision, of generously imagining her. You revise your reader up, in your imagination, with every pass. You keep saying to yourself: “No, she’s smarter than that. Don’t dishonour her with that lazy prose or that easy notion.”

And in revising your reader up, you revise yourself up too.

CSS Color Picker Now Available As A Figma Plugin - Adam Argyle

2026-05-11T15:19:08.000Z

You can now use my <color-input> as a Figma plugin:

HDR Color Picker

Misery is not wisdom - Robin Rendle

2026-05-10T16:34:03.000Z

I might be botching a Hank Green quote with the title of this post but I’ve noticed how it’s increasingly difficult to stay buoyant, to keep light on my toes, to feel optimistic about the future. In Italo Calvino’s Six Memos for the Next Millenium, lightness is the first quality he notes for aspiring writers but I’ve always taken it one step further: I believe that lightness is an essential quality to strive for outside of books and e-essays, too. It requires constantly fighting your own brain-goo, the overwhelming complacency and laziness and shortsightedness of so many, and that eternal memento-mori-feeling that all things eventually turn to muck.

Lightness is heroic! It is courageous! Not to turn a blind eye to terrible things or ignoring them completely, of course, but rather in the face of so much misery to come out hopeful on the other side of it.

So: here are some things that have kept me light.

The new (and final) XOXO website is fantastic. I loved XO. I went three times and I miss it already. Plus, the design of this website sure is something else.

(Although it’s not really surprising because you could always trust the XOXO crew to make one hell of a website.)

Speaking of websites, GT Mechanik is likewise fantastic. With each new GrilliType website I feel a wave of embarrassment—total and utter shame—as I realize how limited my imagination and ambition have become. My favorite part of this website is how you can “tune” to the right scroll destination at the bottom like a radio because usually when someone messes with the scroll it’s a terrible idea—but this works! Sometimes terrible ideas can actually be punk!

Anyway, GrilliType websites are humbling in the best possible way.

Widow’s Bay on AppleTV. So. Very. Good. It’s a dark comedy with light hints of horror. Matthew Rhys is one of the few people who can cuss and make me laugh every time. But I still minimize the window when the scary parts happen.

I don’t believe that all, or even most, of life can be recorded in a database. But sometimes I like the attempt! Sometimes the effort of trying to computerize the world is foolish yet impressive! And perhaps databases don’t need to be filled to the brim with useless analytics or eye tracking junk or keyboard capture malware! Did you know that some databases are filled with trees?

This is why I like the San Francisco Street Tree Map.

Calibre’s Tricklemore Sea has been on repeat all week. It’s brooding, moody, and hypnotic. I’ve been a fan of Dominick Martin’s Calibre for what feels like centuries now and I love how his work yo-yos between drum and bass and something like Tricklemore Sea — lighter, calmer, but with real zoom, a consistent momentum all the way through.

I loved this 7 hour set by Daphni/Caribou’s Dan Snaith, too.

Jenny Volvovski’s website and the Unsolicited section is such a wonderful idea it hurts:

In 2012, I really wanted to design book covers but didn’t have any book cover work. So I hired myself to redesign my personal library. To make the series feel cohesive (and to give myself some structure), I set some rules: green, black, and white for colors; Futura, typewriter, handwriting (and sometimes Caslon Italic) for type; hand-made images whenever possible. As time went on, I started breaking my own rules, and as I was hired to make real covers, I made fewer and fewer unsolicited ones.

The work here is just lovely but this website also fights the eternal fight of being a thing whilst you’re doing a thing. No one will ever give you permission to be a designer or developer, a “real” writer or a “professional” musician. I waited years for someone to tell me what I am, to give me the authorial nod that would let me ascend into doing “real” work. But you have to claim that label for yourself! You doing the work is enough! Ascend!

Also: my second favorite thing about Jenny’s website is the Published section where she shows the options her clients didn’t pick for the final design. Take The Field Guide to Falling Ill for instance. I could look at scrapped book design ideas all day long.

I just finished Saros, Housemarque’s latest rogue-lite mind-bending bullet-hell game. I loved the vague storytelling, the sci-fi mystery, the angsty moodiness of the game. And I loved the ending! Make sure to play beyond the first credit sequence though as the story continues beyond them and wraps up things in a much more satisfying way.

At first I wasn’t sold on the protagonist, Arjun Devraj. I thought he was a selfish jerk. But over the course of the game I fell in love. Especially towards the end, and during the secret ending, where he gets one last shot to make things right.

On that note, I also watched Returnal is a Hell of Our Own Creation which is all about Housemarque’s previous game that I bounced off after the second boss beat me into oblivion. I would watch out though because this video is a total and complete bummer! But it goes into why Housemarque games are interesting, simply because they don’t have satisfying endings. Their characters are all trapped in hell, incapable of escape even after they think they’ve won.

What makes me excited about this is just how much storytelling room there is to play with in games like these. Rogue-lites are all about learning and death, but Housemarque takes that fun gameplay loop and makes it heartbreaking and tragic.

(Sometimes heartbreak can be inspiring!)

I’ve also been playing Mixtape. And goddamn. I worried at first that this game would be too Wes Andersen, too nostalgic or cheesy. And although it is extremely nostalgic it does that end-of-summer-high-school-goodbye thing in the best way possible. The game is built around a mixtape, a playlist, of the last night in your hometown. And as someone who is partly obsessed with making playlists—here’s April 2026—I loved navigating this town, these friendships, these songs.

On Podrocket About Why AI Sucks At Frontend - Adam Argyle

2026-05-07T15:20:36.000Z

Was on PodRocket to chat about my post Why AI Sucks At Front End:

Watch it here or listen

Open web vs AI: what can W3C do? - Hidde's blog

2026-04-28T00:00:00.000Z

At last week’s W3C Advisory Committee meeting, I ran a breakout session on what to do about threats to the open web. We had an interesting conversation. Many interesting points were raised, and some disagreed the web needs saving at all.

Disclosure: as the breakout was part of a Member-confidential event, I generalised and synthesised what was raised, and left out attributions to specific individuals, except where I explicitly asked permission (Members can view the minutes).

Wait, what threats?

Indeed, this session was inspired by earlier sessions at W3C and IETF. I’m not Mark Nottingham, but I really wanted this breakout to happen and discuss possible ways forward in the context of W3C, so I would just go and propose it.

Let’s first consider the problem at hand. A major reason for the web’s success is that it is open in many ways: it’s powered by open standards, it’s open for everyone to publish on (for fun, fame and/or profit), and it’s open for everyone to read. Perfect.

However… this “everyone” includes LLM crawlers, whose visits have been compared to DDOS attacks and frustrate publishers on the web, much more so than search engines have done for years. It causes websites from your average personal blog to Wikipedia struggle with server management and costs.

At the same time, content is increasingly consumed via LLMs and AI agents, which worries websites that care about content quality (like news organisations, governments, or public health information providers) and advertising income.

Comments from the blogosphere

On their blogs, various people added their thoughts to this discussion. Mike Masnick wrote about using AI assistants to create our own tools, and provide them with consent driven contexts via AT Proto. In Endgame for the open web, Anil Dash said we should remember how much the open web matters, volunteer at and support organisations that fight for it, like EFF and Mozilla, and consider what sort of policy or regulation could work. Julien Genestoux added we’re killing the web and recommended individuals to “publish in places you can leave”, meaning places where you can export data so that you can easily take it elsewhere. He also urged the community to rebuild social primitives, like identity, follow graphs, payment systems and discovery mechanisms, because existing primitives are provided by companies whose values don’t align with ours. Ben Werdmuller responded agreeing with Julien’s points, suspecting there’s interest in alternatives, and we need to engage one another more: “it is a work of engaged citizenry that verges on activism”.

Value extraction and end users

Chris Needham of the BBC takes part in the AI Preferences work at IETF and shared his worry about the value extraction that is going on, by AI companies, from companies that make the content, saying we must ensure fair value exchange. In that context, he mentioned the SPUR Coalition as one of the initiatives that will develop standards for responsible use of journalist’s content.

Not everbody agreed that the web “needs saving”, as I phrased it in my introduction, or that the problem is new: search engines have always scraped content from the web, in order to help users find that content. And we shouldn’t forget that a lot of end users want agents to help them browse the web, various attendees said. They can helps make sense of the enormous collection of information that is today’s web, in which it is increasingly hard to find stuff. Might AI agents may just qualify as Web User Agents?

What could W3C do?

In addition to AI Preferences (IETF) and standards for responsible use of journalistic content (SPUR Coalition), these are some things participants suggested W3C can do:

Ensure AI companies are engaged in our standards work and community, possibly through workshops that are relevant for them (this topic is, as a web where nobody makes content will drive them out of business too);
Building on that: serve as a bridge between AI companies and content creators… this is about publishers like the ones that are active at the W3C, but also much broader, including smaller content creators like bloggers or influencers;
Take threats to the open web into account in work on WebMCP and the “agentic web”, making it an “open agentic web”, built on open standards and technologies (rather than proprietary) and based on open principles (rather than walled gardens);
Consider a push model for content, so that websites don’t need to be crawled every second;
Think about standards for payments (Web Monetization?) and attribution, complementing what SPUR will work on.

Summing up

I was glad to see this breakout full, and though we weren’t able to close out the queue before we ran out of time, we got a little closer to thinking about possible strategies. I am sure this conversation is going to continue in the next month, inside and outside of the W3C community. If you're reading this and have thoughts, do feel free to share!

Originally posted as Open web vs AI: what can W3C do? on Hidde's blog.

Reply via email

CSS Recently In All Browsers - Adam Argyle

2026-04-26T01:32:25.000Z

Here's some rad CSS that just hit baseline between October 2025 - April 2026.

At this point with the following features, the "I can't use it" vibe can shed away.

Anchor Positioning #

Anchor positioning lets you natively tether components to target elements without adjusting DOM semantics or crowding the main thread.

This widget doesn't show green across the board as of writing this, there are sub features that aren't all supported, but the general featureset is ready.

Checkout Tab's talk about anchor called Anchor Positioning at CSS Day 2025.

Checkout my post on AIM if you missed it, neat anchor technique for FLIP-like transitions with anchor.

@scope #

CSS selector scoping, not style scoping; a common misconception.

However! With @scope we can simplify naming conventions or robobarf classnames just to avoid global cascade collisions.

The "donut" feature is very special, it can limit styles from cascading into nested components by setting an end to the selector (a donut hole 🍩).

Chris gave a talk about @scope called Scope In CSS at CSS Day 2025.

Name Only Container Queries #

@container no longer requires a size condition.

You can now conditionally style elements just by name:

.sidebar {
  container-name: sidebar;
}

.card {
  display: grid;

  @container sidebar {
    grid-auto-flow: column;
  }
}

Chrome 130+
Safari 26.4
Firefox 149

Chris wrote a good post about it over on FrontEnd Masters.

shape() #

Responsive, native CSS geometry.

Draw complex clipping paths using standard CSS syntax and dynamic units (like rem or calc()) instead of being locked into rigid SVG pixel coordinates.

Temani wrote a rad post about it over on FrontEnd Masters and a "complex" post CSS-Tricks.

shape-outside with xywh() and rect() #

Wrap inline text around precise geometric boundaries with xywh() and rect().

Typographical control over floating text flows without relying on images or clip-paths.

Chrome 128 (recent)
Safari 18
Firefox 149

Still no pretext, but also still coo.

light-dark() with image #

Dark mode backgrounds in a one liner.

Pass <image> values (like url() file paths or linear-gradient) directly into light-dark().

Chrome 147
Safari 26.4
Firefox 150

Bramus wrote a post about it over on his blog.

view-transition-class and Types #

SPA-like routing animations.

With view-transition-class you can target massive collections of DOM nodes with a single animation rule, while JS View Transition Types let you programmatically direct "forward" or "backward" contextual motion.

rcap, rch, rex, ric #

Even more typographic precision.

I wrote about these here.

Uncertainty is healthy - Robin Rendle

2026-04-25T16:15:50.000Z

Entangled Life by Merlin Sheldrake is a fantastic book all about mushrooms and fungi but what made it a breezy read for me was how weird they are. It’s like Merlin got his hands on creatures from another dimension where the rules of biology simply don’t apply.

First: Merlin goes into great depth about genes and how they don’t have to be passed on from parent to child. Some organisms can inherit traits horizontally, or from creature to creature. Imagine if you bumped into Hans Zimmer and now you can play the piano. Weird!

Second: Merlin has a whole chapter about lichens which messed me up. Lichens are odd because they’re symbiotes, Venom-style, and have a curious relationship with fungi. In fact, the closer you look at them the harder it is to see individual creatures. When you look closely at “a” lichen you see ten thousand other organisms as if you were looking at a city, a collective, a thing of multiple organisms that we just pretend is an individual because that’s how we see ourselves. In fact, when looking at mushrooms, thinking about the world in this way — of separate individuals — doesn’t make much sense at all.

All of this is gloriously sci-fi, as Merlin writes:

To talk about individuals made no sense anymore. Biology—the study of the relationships between living organisms—had transformed into ecology—the study of the relationships between living organisms.

This reminds me of James Burke’s Connections where he argues that what we think of as separate disciplines — like farming and math or subjects like biology and web design — aren’t separate things. Every discipline is a big soupy mess that gets blurrier and blurrier the closer you look at them.

In the same way, mushrooms and fungi don’t care about conforming to our world of individual identities:

The study of relationships can be confusing. Almost all are ambiguous. Have leaf-cutter ants domesticated the fungus that they depend on, or has the fungus domesticated the ants? Do plants farm the mycorrhizal fungi that they live with, or do the fungi farm the plants? Which way does the arrow point? This uncertainty is healthy.

I love this, of course. It feels optimistic, hopeful, sci-fi, and Star-Trek-as-hell, to see ourselves as part of a collective instead of individuals. To see that collaboration is strategic and necessary.

Cozy Ux - Adam Argyle

2026-04-21T02:29:26.000Z

I love cozy games. I want more cozy software.

It's such a great UI/UX goal: chill, empowerment, and joy.

Cute Ass Whiskey Coaster - Adam Argyle

2026-04-17T20:35:16.000Z

So stinkin' cute!

WWW Ep240 AI Cowgorithm Disrupts The Herd - Adam Argyle

2026-04-16T22:57:10.000Z

Ep #240
AI Cowgorithm Disrupts The Herd
(presented by Warp)

🕹️ Token spending
🧐 Autoresearch
🌗 Day Shift / Night Shift
🐄 Cowgorithms
💻 Terminal tools
🥃 Whiskey & whatnot

⤷ whiskey.fm · youtube · spotify · apple

Why AI Sucks At Front End - Adam Argyle

2026-04-12T05:48:57.000Z

and nerdy.dev written on it." height="674" width="1920" />

AI is a sycophantic dev wannabe that skimmed a shitload of tutorials. You get the results of a probabilistic guess based on patterns it saw during training. What did it train on? Ancient solutions, unoriginal UI patterns, and watered down junk.

I'm about to rant about how this is both useful and lame.

The Good #

AI loves the boring stuff. It thrives on mediocrity.

If you want some gloriously unoriginal UI, it has your back 😜

Scaffolding: Generic regurgitation of patterns it's seen, done.
Tokens: Migrating tokens or mapping them out? It eats this tedious garbage for breakfast.
Outlining features: Generic lists ✅
Lying to your face: Confident hot garbage on a silver platter. It'll hand you a snippet, dust off its digital hands, and tell you it finished the work. It did not finish the work.

Aka: If it's a well-worn pattern, AI is there to help you copy-paste faster. Which, for a lot of programming, is totally the case. I'm genuinely finding a lot of helpful stuff in this department.

The Bad #

Pixel perfection & bespoke solutions… what are those?

The exact second you step off the paved road of unoriginality, it faceplants.

Bespoke solutions & custom interactions: Try asking it for some scroll-driven animations or custom micro-interactions. It will invent a CSS syntax that hasn't existed since IE6.
Layout & Spacing: Predicting intrinsic/extrinsic page properties? It's already bad at math, how could it get this rediculously dynamic calculation correct. Spacing? Ha, seems reasonably to expect symmetry, but it's terrible at the math.
Combined states: Pinpointing where to edit a complex component state makes it cry.
Accessibility: It throws aria-hidden="true" at a wall and hopes it sticks.
Performance: It will give you the heaviest, jankiest solution unless you explicitly ask it to be for a specific (apparently "indie") performance solution.
Tests: Writing good tests? Good, no. A lot, yes.

And the absolute best part? The more complex the component gets, the slower and dumber the front-end help becomes. Incredible how it can one shot a totally decent front-end design or component, than choke on a follow up request. Speaks to what it's good at.

Why? #

1. It trained on ancient garbage #

It lacks modern training data.

It has an excessive reliance on standard templates because that's what the internet is full of. Modern CSS? It's barely aware of it.

2. It literally cannot see #

It's an LLM, not a rendering engine!

It's notoriously bad at math, and throwing screenshots at it means very little. It's stabbing in the dark.

This leads to the classic UI interaction:

AI: "I'm done! Here is your perfectly crafted UI."
Me: "There's a gaping hole where the icon should be, fix the missing icon."
AI: "You're absolutely right. Let me fix that for you."

3. It doesn't know WHY we do things #

It doesn't understand the "why" behind our architectural decisions.

SDD, BDD, or state machines might help guide it, but the models weren't exactly trained on those paired with stellar solutions.

We're asking a giant text-predictor to make new connections on the fly. We can get it there, but there's so much to consider we have to spell it out before it starts making the connections we want.

4. Zero environmental control #

It doesn't control where the code lives.

It can write annoyingly amazing Rust, TypeScript or Python, but those have the distinct advantage of a predictable (pinnable!!! like v14.4) environment the code executes in.

That's not how HTML or CSS work, there is no pinning the browser type, browser window size, browser version, the users input type (keyboard, mouse, touch, voice), their user preferences, etc. That's complex end environment shit.

The list goes on too, for scenarios, contexts and variables the rendering engine juggles before resolving the final output. The LLM doesn't control these, so it ignores them until you make them relevant.

Even prompting in logical properties, you have to ask for this kind of CSS. These should be CSS tablestakes output from LLMs, but it's not. And even when you ask for it, or provide documentation that spells it out, it's not guaranteed to work.

The place where HTML and CSS have to render is chaotic. It's a browser, with a million different versions, a million different ways to render, a million different ways to interact with it, and a million different ways to break it.

It's a moving target, and LLMs are terrible at moving targets.

Damnit humans #

We're a LLM combinatorial explosion.

We're wildly unpredictable targets. We change our minds, we switch viewports, we change theme preferences, we changes devices, we change browsers, we change browser versions, we switch inputs, we change our everything.

We're not a static target. We're not a pattern that can be learned.

There is a "human mainstream" of behaviors, preferences, and expectations where LLMs can be genuinely helpful; but our "full potential" matrix will be exploding LLM output patterns for a long time to come. IMO at least.

unless we Borg.

WWW Ep239 Flogging Margins With The Dropcap Murphys - Adam Argyle

2026-04-10T16:45:12.000Z

Ep #239
Flogging Margins With The Dropcap Murphys
featuring Dave Rupert and Chris Coyier

™️ Brand Truth vs Brand Slop
🔫 One-Shot Apps
🧑‍🎨 CSS in the Age of AI
🪾 Context Rot
🧛🏻‍♂️ Energy Vampires & AI Brain Fry
🥃 Whiskey & whatnot

⤷ whiskey.fm · youtube · spotify · apple

Migration To Fresh 2 And Denos Latest Console Complete - Adam Argyle

2026-04-03T19:16:36.000Z

dash.deno.com → console.deno.com
Fresh → Fresh 2

This was a large chunk of work 😅

Please report any regressions.

WWW Ep238 Nextjs Is Dead Long Live Nextjs Presented By Warp - Adam Argyle

2026-04-02T16:10:02.000Z

Ep #238
NextJS is Dead, Long Live NextJS (presented by Warp)

🏭 Dark Factories
🤖 YOLO mode
❌ T shape vs X shape
🦞 OpenClaw
💀 Death of SaaS
🥃 Redwood Empire & whatnot

⤷ whiskey.fm · youtube · spotify · apple

Notes from “AI & the Future of Sustainability Reporting” - Hidde's blog

2026-04-01T00:00:00.000Z

Today I was at an event on sustainability reporting. It was hosted by Digital Catapult and excellently moderated by Jo Lindsay Walton and Chanell Daniels.

I was glad I was able to to join, given my interest in AI ethics (at work and W3C), a greener future (as a contributor to WSG), and in reporting methodologies (but for accessibility, as an editor of WCAG-EM). I don't know a lot about sustainability reporting, so was keen to learn.

The event had a keynote, two panels and two interactive session. In this post, I'll share notes and links that I picked up from the day. And yes, this post too adheres to my AI statement.

The event was backdropped by most of London's St Pancras Station.

Keynote: why reporting matters

1.5 degree promise

In his keynote, Jo Lindsay Walton positioned the event as being about climate change more than AI, reminding us that we promised one another to stay “well below 2 degrees Celcius” compared to pre-industrial levels, and recently “effectively missed the 1.5 degrees target”. This is why carbon removal matters, now more than ever.

The bath tub

Using John Sterman's bathtub metaphor, he said we're collectively pouring more water into the tub than we're allowing to drain. “We need to equalise the drain with the tab going into the bath tub”, Jo explained, “doing more removals than emissions”. This applies to AI in the sense that AI is the main reason that tech companies are currently increasing (rather than decreasing) their energy use, they're pouring more into the tub than they are letting out. The assumption may be that technology could eventually lead to reaching carbon reduction goals faster, but the required scale and pace is not yet to be seen. At the same time, AI can divert our focus from greenhouse gas removals.

Policy implications

The report “Big Tech’s climate performance and policy implications for the UK”, that Jo mentioned, notes “there is a real risk that emissions from the AI build-out will outstrip any climate gains as tech companies abandon net zero goals and pursue huge AI-driven profits”.

Net benefit/detriment framing

“Is AI a net benefit or net detriment to sustainability?”, people often ask. Jo explained this is the wrong question—it's misleading as it conflates many different kinds of AI systems that have different infrastructural requirements. It puts things like ChatGPT in the same bucket as other more traditional systems. Like when we talk transport we don't want to conflate planes with bicycles. Jo and colleagues reject this framing in their paper “Modelling diverse futures of AI and the climate”, recommending an approach that includes open data, being explicit about what's uncertain, and inclusion of broad stakeholders.

How you measure matters

Measuring sustainability is essential, and there is a risk of divergence between scores (more on that below). Without good measurements, claims could become greenwashing (a practice EU Directive 2024/825 effectively prohibits). We mention greenwashing in the introduction of the Web Sustainability Guidelines, and in the group making WCAG, the accessibility standard, something similar has our attention: it's important that people can't “abuse” the system.

On greenwashing, Jo mentioned “The AI Climate Hoax: Behind the Curtain of How Big Tech Greenwashes Impacts” (PDF), that talks about how being vague about the meaning of the word “AI” helps corporations avoid responsibility.

Differences in methodologies

Lastly, Jo discussed that between ways to measure carbon, results can differ quite a bit, even between methods that are considered reasonable. A (preprint) paper explaining this effect is Beyond Counting Carbon: AI Environmental Assessments Struggle to Inform Net Impact Decisions.

Panel 1: Is AI sustainable?

After Jo's talk, he moderated a panel on risks, trade-offs, and futures.

The supply-driven nature

Loïc Lannelongue said it's problematic that AI is pretty much entirely supply driven, rather than driven by demand to solve problems. Copilot, for example, is introduced absolutely everywhere and it's like we're waiting for someone to raise their hand to find reasons to use it… that's terrible from a sustainability perspective.

The full lifecycle

Melissa Gregg talked about how helpful it was to, when she worked at Intel, bring the company's sustainability folk together with engineers, to close specifically those gaps. She also explained that it was important to move from spend-based carbon accounting to a more holistic type of accounting that includes embodied carbon and the full lifecycle. Especially as companies release devices that contain more and more chips (like smarts glasses).

Accuracy

The panel discussed whether accuracy in reporting even matters, as even inaccurate reporting can already help movement in the right directions. Besides, there is also sustainability impact (concerning the S of ESG) that cannot easily be captured by metrics, like the effects of increasing digitalisation on working conditions, human rights, and communities.

Policy and five 9's

Policy and regulation was also discussed. Loic mentioned that it's common for data centres to guarantee uptime with “five 9's”, meaning they guarantee to be available 99,99999% of the time. This sounds great and very reliable indeed, but in terms of energy it means that those data centres use diesel power to achieve it. We could introduce policy and regulation for such metrics, and decide when it really matters: a cat video on YouTube could be unavailable for a couple of minutes, while for a hospital, the possible life and death consequences of downtime could justify lots of 9s. With most more services striving for less 9s, might we avoid lots of carbon emissions?

Panel 2: AI tools in sustainability reporting

The second panel was about using AI in the process of sustainability reporting.

I was surprised to hear that many in the panel advocated for using AI in all sorts of ways, including for analysis. Given what we know about hallucination and the inner workings of LLMs (and then I'm still ignoring the energy use), I struggle to understand this cost/benefit analysis. But admittedly, I don't write sustainability reports, so I can't really speak to whether my choice deprives me of productivity wins that would have enabled me to do more carbon removal.

AI helps do sustainability work faster, was a common theme in this panel. One of the panelists explained his agentic AI setup as “having cognitive slaves (sic) work as assistants for you”. I struggled to identify with the desire to “have” “slaves” (I could not separate it from my immediate connotation with a historical wrong), and felt attributing “cognition” to agents anthropomorphises them. There is way too much that philosophers, psychologists, and neuroscientists don't yet understand about cognition to back that up.

In this panel, it was interesting to hear how the panelists worked with AI, and get their insights on how AI benefits their sustainability work, as well as how it may end up impacting the industry.

Activity: the future

In the last session we worked in groups to imagine the year 2040, when all the problems with AI have disappeared and all is peachy. Jo acted as a time machine that took us from 2040 to 2035 to 2030, having us figure out what changes to make to get to this future.

This was a fun thought process, and it had me considering what it would be like to have public-interest, non-problematic AI only. One group proposed to bring critical thinking to the classroom, as early as possible. As someone who was lucky enough to have dedicated philosophy classes in my high school curriculum, that I still often think about decades later, I wholeheartedly agreed.

Summing up

I had a great time at the “AI and the future of sustainability reporting” event, and am grateful to the organisers for putting it on, and Digital Catapult for hosting it in their beautiful location.

Originally posted as Notes from “AI & the Future of Sustainability Reporting” on Hidde's blog.

Reply via email

Software is a feeling - Robin Rendle

2026-03-29T18:52:09.000Z

I made a few updates over the weekend!

All the images in /photos are smartly resized and they also load as you scroll so I’m not sending you 700GB of data down the wire. I have a lot of ideas about what I’d like to turn /photos into but either way it’s amazing how little work is required to build your own personal Flickr with a bit of CSS and hooking up a folder of images to 11ty (now Build Awesome).

I need to figure out how I can have a nice folder of images and make it easy to add more and have alt text with a tiny description of the location or the scene though. But as any blogger will tell you: the greatest threat to any blog is how difficult it is to maintain and update. Improving the accessibility and design is great but it can’t come at the expense of adding more faff to update things. That’s how all blogs eventually die.

Oh, also? I should definitely add a feed for /photos.

I mentioned Mass-Driver’s MD UI last week and since it was designed to pair so well with MD IO—the typeface I use for body copy here—I decided to make the switch. Now MD UI is the typeface I’m using for navigation and titles and for the first time I’m using a variable font which feels nice and future proof if I want to tinker with the design.

You can see it in the sidebar and it’s just great. But I particularly love, love, love MD UI at big sizes. It packs a real punch and I’ve spent the last twenty minutes looking at the type specimen and thinking about how I can use it elsewhere in my website besides the sidebar. I’ll think about it more!

Unrelated note: I love the italics of MD IO and I went back and forth making the whole site in italics a while ago because I love it so much. Look! Isn’t this just great? It has so much style and quirkiness. This is why I love to quote books and websites because italics make all writing better and smarter and hotter. Monospace italics ftw!

A while back I noticed a strange flash of text as I navigated between pages. After tinkering around it looks like the html/body element needs a color background when you’re using view transitions. After that tiny fix things certainly feel a lot more like a native IDE but my website still feels...fragile? And certainly not native enough for my liking.

I wonder how else I can make the interface more stable, more predictable, and less like a haphazard pile of links and text. (Generally I hated the curse of the 2010s, the React-ification of the world wide web, and always felt that technology greatly contributed to the way that most websites were hobbled with bloated, junk technology. But I think the general desire from folks to want stable-feeling websites, to make their websites feel like “grown-up” or “stable”, was correct. It should’ve just been done in a standards-compliant-way instead of a bloated-terrible-mess kind of way. Anyway!)

It’s been ~3 months since I launched this variation of my website and it feels like this one might last all year. I can’t remember if that’s happened before! Most designs last a couple of months until I anxiously tear them to shreds. But this guy? I still like it! I have no grand desire to redesign everything but only to tweak, tweak, tweak.

I have this idea for a “collection”—tagged posts, basically—where I can group notes and essays into playlists that have a common theme, a suggested method of entry. I think Paul Ford did this in a previous design of Ftrain which I always admired. Right now I have a chronological list of posts but there’s other ways to slice them up, other ways to present them, that I’ve mostly left ignored.

The only big bug that bugs me right now is how selecting a folder in the sidebar takes you to a full screen page of the contents. So you’re viewing the folder’s contents in the sidebar and the main content area at the same time. I thought this was a good idea a while ago so that I can provide more information about a post but now it feels finicky and annoying to open a folder and navigate.

Perhaps this is why the interface feels a little untrustworthy, a little unpredictable at times. Software is a feeling and sometimes that feeling is hard to understand, difficult to parse. Investigating what makes software feel “correct” is using it ten thousand times for months on end, and what was once a punk rock idea in a drawing app suddenly feels slow and wrong and weird when transmuted into code and brought into life.

WWW Ep237 Using Ai Wrong With Leon Noel And Danny Thompson - Adam Argyle

2026-03-26T15:49:22.000Z

Ep #237
The Transactional Trap:
How 97% of Developers Are Using AI Wrong
- with Leon Noel & Danny Thompson

🤖 agentic dev
💬 prompt engineering
🦞 OpenClaw
🎼 orchestration, harnesses, and models
💀 leet code interviews
🕹️ Pokemon Go
🥃 whiskey & whatnot

⤷ whiskey.fm · youtube · spotify · apple

Rfc Latest Color Input Concept - Adam Argyle

2026-03-23T05:03:51.000Z

Try the latest <color-input> concept
(I think it's hella rad)

⚓️ anchor() progressive enhancement
🤖 workers for color compute
🙈 new animations & interactions
🎨 gamut boundaries, mapping, stretching
☯️ contrast scores
& much more

QA and comment on GitHub if you may.

Landslide; a ghost story - Robin Rendle

2026-03-21T16:24:19.000Z

Erin Kissane:

There is a lot—like a lot a lot—of research suggesting that using social media to keep up with news does not actually inform us but does makes us feel more informed. Many studies of many populations have found for knowledge about what’s happening to us, social media essentially magnifies the Dunning-Kruger effect. When we talk about brainrot, we’re not wrong.

This mirrors how I’ve felt for a long while. Mostly, guilty about taking a big step back from reading or listening to the news. It’s too difficult. And the more I read, the less informed, the less connected to the world, I feel. Reading the news in 2026 feels like connecting to a hive mind of 8 billion people screaming in every direction. I want to be a part of it, I want to contribute, but I feel weak under its influence. It almost feels like knowing the world in this way doesn’t feel useful or practical or even possible, emotionally. Thankfully the data backs me up but it doesn’t feel great, huh.

Erin also writes:

Real belief in collective knowledge may also suggest that caring for our own knowledge formation in social systems is a way of caring for the system as a whole. Necessary, if not sufficient.

I like this. Caring for my dinky garden blog is how I show care for the overwhelming, unknowable, impossible mass that is all human knowledge.

If I can be a steward over here, you can be a steward over there…

Reading without reading - Robin Rendle

2026-03-20T05:30:10.000Z

Mass-Driver just released MD UI, a neo-grotesque type family that’s been designed to compliment Rutherford Craze and Co.’s already excellent mono-space. And just as the name implies, MD UI is designed for a different kind of reading:

When we think about reading, we tend to picture books, or magazines, or business reports. Blocks of text, traversed from beginning to end. There’s another kind of reading, though, one which I think we do much more often without really noticing it.

It’s a kind of instantaneous scanning of a word or two at a time: the hour on a smartwatch, or the label on microwave dial, or the URL in your web browser. I like to call this interfacing, because it’s less about digesting information (or experiencing the text) than about interacting with your immediate surroundings. You don’t think of it as reading at all — it just happens.

Interfacing is a great word! I realize now just how limiting it is to describe all types of reading as simply reading, as if all flowers are flowers or all birds are birds. The world is more complicated! There is nuance here! And of course the way we process information is more complex than a single word can hope to describe.

Also, I appreciate the confidence from Mass-Driver in this new type release. There’s a self-assuredness (without ego!) that can be found in the design notes that I truly admire:

MD UI isn’t so much a revival of the International Style as an anticipation of its potential future. It’s an attempt to take the familiar and push it just one step further; to fix a couple of the annoyances while leaving what works alone. That’s easier said than done. But it’s done.