Maggie Appleton
Design engineer and illustrator; makes visual essays on programming, anthropology and what language models do to the way people write.
Maggie Appleton did not write this page. ppll compiled it from their own words, each linked to the source. Is this you? · Report a mistake
Latest Site
What they believe
-
Interfaces between humans and agents are too thin, with little translation between what humans see and what agents see.
Another way to put this is that our interfaces between humans and agents are very thin. There's not a lot of translation between what humans see and what agents see.
maggieappleton.comSaid 14 Sept 2026This on korrents
-
There is plenty of work that humans are needed for and robots cannot do, but not enough ability or willingness to pay people to do it.
There is no shortage of physical, social, educational, and cultural work that we need humans to do for one another that cannot, in any wild imagining of agent capabilities, be done by our current or near-future robots. We just can't figure out how we would pay people enough to do these jobs.
maggieappleton.comSaid 18 Aug 2026This on korrents
The bold line is korrents’ wording; the quote is theirs. These 2: what they wrote before what they said aloud, one per source, newest first.
Feed
As its own page →Hiding
4 November
1 November
29 October
28 October
26 October
24 October
23 October
22 October
20 October
19 October
8 October
7 October
6 October
14 September
From one piece Planning with Agents: Divided Worlds, Boundary Objects, and Thicker Interfaces 3 beliefs · maggieappleton.com
-
korrents.com
A plan handed to an agent should already have made meaningful contact with the important parts of reality.Their words
By the time we hand the plan over, it should already have made meaningful contact with the important parts of reality.
-
Their words
The agent is not the prioritised party in this interaction. We are. The agent should be doing far more labour for human benefit.
-
Their words
Another way to put this is that our interfaces between humans and agents are very thin. There's not a lot of translation between what humans see and what agents see.
18 August
From one piece August 2026 2 beliefs · maggieappleton.com
-
Their words
I am genuinely hopeful agents might put a dent in this problem, in the same way washing machines and dishwashers did.
-
Their words
There is no shortage of physical, social, educational, and cultural work that we need humans to do for one another that cannot, in any wild imagining of agent capabilities, be done by our current or near-future robots. We just can't figure out how we would pay people enough to do these jobs.
1 August
-
Lovedrcmnd.app
YesteryearTheir words
Yesteryear and loved it, just like everyone else. Strong recommend.
18 July
From one piece In Memoriam, Encountering World War I, and Thinkable Horrors 3 beliefs · maggieappleton.com
-
Their words
It reminded me we are not as far as we like to imagine from mass, global, industrial-scale death. Just over a hundred years past it. Likely only a few decades from it.
-
Their words
I considered how quickly we forget this level of suffering is possible. How easily it becomes acceptable.
-
Their words
Fiction can often tell a more comprehensive and representative story of history than any single "true" account by stitching together many people's lived experiences, then zooming out to explain the context across multiple characters.
13 April
From one piece One Developer, Two Dozen Agents, Zero Alignment 3 beliefs · maggieappleton.com
-
korrents.com
Pull requests and issues are the wrong primitives for the speed, shape and volume of agentic software work.Their words
The PR and the issue are the wrong primitives that can't handle the speed, shape, and volume of agentic work.
One Developer, Two Dozen Agents, Zero Alignmentmaggieappleton.com
-
korrents.com
More individual output does not solve problems that need a team to communicate and coordinate; it makes them worse.Their words
More individual output doesn't solve problems that require everyone to communicate and coordinate. It makes them worse.
One Developer, Two Dozen Agents, Zero Alignmentmaggieappleton.com
-
korrents.com
With implementation cheap, deciding what to build is the hard question in software, not how to build it.Their words
The hard question is no longer how to build it. It's should we build it. Agreeing on what to build is the new bottleneck.
One Developer, Two Dozen Agents, Zero Alignmentmaggieappleton.com
23 January
2 January
5 August 2025
2 August 2025
25 May 2025
24 March 2025
13 March 2025
5 March 2025
20 February 2025
26 January 2025
Nothing matches.
Beliefs
Korrents What they believe 20 beliefs — each backed by an exact quote.
Each is a — compiled by korrents.com, not by them: the one-line wordings are korrents', the quotes are theirs.
Recent
Interfaces between humans and agents are too thin, with little translation between what humans see and what agents see.
Another way to put this is that our interfaces between humans and agents are very thin. There's not a lot of translation between what humans see and what agents see.
Planning with Agents: Divided Worlds, Boundary Objects, and Thicker Interfaces Said 14 Sept 2026
In human-agent interaction the human is the prioritised party, and the agent should do far more labour for the human's benefit.
The agent is not the prioritised party in this interaction. We are. The agent should be doing far more labour for human benefit.
Planning with Agents: Divided Worlds, Boundary Objects, and Thicker Interfaces Said 14 Sept 2026
A plan handed to an agent should already have made meaningful contact with the important parts of reality.
By the time we hand the plan over, it should already have made meaningful contact with the important parts of reality.
Planning with Agents: Divided Worlds, Boundary Objects, and Thicker Interfaces Said 14 Sept 2026
Show 17 more
There is plenty of work that humans are needed for and robots cannot do, but not enough ability or willingness to pay people to do it.
There is no shortage of physical, social, educational, and cultural work that we need humans to do for one another that cannot, in any wild imagining of agent capabilities, be done by our current or near-future robots. We just can't figure out how we would pay people enough to do these jobs.
August 2026 Said 18 Aug 2026
Always-on AI agents might put a dent in the burden of unpaid household mental labour, the way washing machines and dishwashers did.
I am genuinely hopeful agents might put a dent in this problem, in the same way washing machines and dishwashers did.
August 2026 Said 18 Aug 2026
Fiction can give a more comprehensive and representative picture of history than any single true account, because it combines many people's lived experiences and explains the wider context.
Fiction can often tell a more comprehensive and representative story of history than any single "true" account by stitching together many people's lived experiences, then zooming out to explain the context across multiple characters.
In Memoriam, Encountering World War I, and Thinkable Horrors Said 18 Jul 2026
People quickly forget that suffering on the scale of the First World War is possible, and it becomes acceptable easily.
I considered how quickly we forget this level of suffering is possible. How easily it becomes acceptable.
In Memoriam, Encountering World War I, and Thinkable Horrors Said 18 Jul 2026
The world is not as far as people like to imagine from mass, global, industrial-scale death, and is likely only a few decades from it.
It reminded me we are not as far as we like to imagine from mass, global, industrial-scale death. Just over a hundred years past it. Likely only a few decades from it.
In Memoriam, Encountering World War I, and Thinkable Horrors Said 18 Jul 2026
With implementation cheap, deciding what to build is the hard question in software, not how to build it.
The hard question is no longer how to build it. It's should we build it. Agreeing on what to build is the new bottleneck.
One Developer, Two Dozen Agents, Zero Alignment Said 13 Apr 2026
More individual output does not solve problems that need a team to communicate and coordinate; it makes them worse.
More individual output doesn't solve problems that require everyone to communicate and coordinate. It makes them worse.
One Developer, Two Dozen Agents, Zero Alignment Said 13 Apr 2026
Pull requests and issues are the wrong primitives for the speed, shape and volume of agentic software work.
The PR and the issue are the wrong primitives that can't handle the speed, shape, and volume of agentic work.
One Developer, Two Dozen Agents, Zero Alignment Said 13 Apr 2026
We are on the verge of a golden age of home-cooked software, made by a new kind of developer: the barefoot developer.
The slightly bold theory I put forward in this talk is that we’re on a verge of a golden age of local, home-cooked software and a new kind of developer – what I’ve called the barefoot developer.
Home-Cooked Software and Barefoot Developers Said 30 May 2024
The web is being flooded with coherent, generic, generated text that buries the writing worth reading.
These new models are poised to flood the web with generic, generated content.
The Dark Forest and Generative AI Said 31 Dec 2022
The volume of generated content arriving is different in kind from the spam the web already had, and we are not prepared for it.
But I think the sheer volume and scale of what’s coming will be meaningfully different. And I think we’re unprepared.
The Dark Forest and Generative AI Said 31 Dec 2022
The way to show a reader you are not a language model is to think critically and originally, because that is the one thing the models cannot yet do.
Easier said than done, but one of the best ways to prove you’re not a predictive language model is to demonstrate critical and sophisticated thinking.
The Dark Forest and Generative AI Said 31 Dec 2022
Once generated text is everywhere, readers will become sceptical that there is a person on the other end of anything they read online.
Every time you find a new favourite blog or Twitter account or Tiktok personality online, you’ll have to ask: Is this really a whole human with a rich and complex life like mine?
The Dark Forest and Generative AI Said 31 Dec 2022
Writing that is specific, local and grounded in a real life is the cheapest available proof that a person wrote it, because models default to generics.
Referencing obscure concepts, friends who are real but not famous, niche interests, and recent events all make you plausibly more human.
The Dark Forest and Generative AI Said 31 Dec 2022
Machine-generated writing raises rather than lowers the bar for people: once a machine can produce a competent summary, there is no value in a person producing one.
In a world of automated intelligence, our goalposts for intelligence will shift. We’ll raise our quality bar for what we expect from humans.
The Dark Forest and Generative AI Said 31 Dec 2022
Language models have codified formal written language at a moment in time, which leaves informal, inventive, in-group speech as the place where humanness still shows.
We have designed a system that automates a standardised way of writing. We have codified la langue at a specific point in time.
The Dark Forest and Generative AI Said 31 Dec 2022
People who want unoptimised, personal conversation have retreated from the open web into closed rooms, and that retreat is a real loss.
As the forest grows darker, noisier, and less human, I expect to invest more time in in-person relationships and communities. And while I love meatspace, this still feels like a loss.
The Dark Forest and Generative AI Said 31 Dec 2022
Beliefs others hold too
The web is being flooded with coherent, generic, generated text that buries the writing worth reading. 3 hold this
These new models are poised to flood the web with generic, generated content.
The Dark Forest and Generative AI Said 31 Dec 2022
What is a korrent?
A korrent is a belief a person has stated in their own words: one sentence stating the claim, backed by a quote and a source, kept at korrents.com.
Under a name here, the quoted block is what they actually said. The korrent beneath it is the claim those words support, in korrents' wording — tap it to see the record, its source, and who else holds it.
Nobody here wrote their own korrents. They are compiled from public statements, and a person can change their mind, which is recorded too.
Recommends rcmnd
rcmnd What they recommend 5 things — each backed by an exact quote.
Compiled by rcmnd.app, not by them: the quotes are theirs, and each links to the post it was copied from.
Recent
Loved Yesteryear
Yesteryear and loved it, just like everyone else. Strong recommend.
maggieappleton.com 1 Aug 2026
Loved Empire of Pain: The Secret History of the Sackler Dynasty
But Radden Keefe is a brilliant writer and his detailed historical narratives make it worth my discomfort.
maggieappleton.com 1 Dec 2024
What others say about Empire of Pain: The Secret History of the Sackler Dynasty →
Loved On Writing: A Memoir of the Craft
On Writing ” – both excellent. The former for grounded advice on non-fiction communication, the latter for an extraordinary account of persistently writing through double-shift minimum wage jobs and a raging alcohol and drug problem.
maggieappleton.com 1 Sept 2023
Show 2 more
Loved The Sense of Style: The Thinking Person's Guide to Writing in the 21st Century
On Writing ” – both excellent. The former for grounded advice on non-fiction communication, the latter for an extraordinary account of persistently writing through double-shift minimum wage jobs and a raging alcohol and drug problem.
maggieappleton.com 1 Sept 2023
Read Working in Public
Reading Working In Public Working in Public by Nadia Eghbal
maggieappleton.com 1 Sept 2020
Where they publish
Site maggieappleton.com Her digital garden: illustrated essays and notes. Has a feed.
Recent
- Fortune Telling and Speculative Design 4 Nov 2026 Practical tools for design engineers to explore possible futures through speculative design
- Planning with Agents: Divided Worlds, Boundary Objects, and Thicker Interfaces 14 Sept 2026 Why we need richer, thicker interfaces and better boundary objects for collaborative planning with agents
- August 2026 18 Aug 2026 I'm writing this from a homely Jacobean-manor-turned-hotel in the West Country, covered in layers of green vines and circled by centenarian trees, on a rare weekend away. Everything is warm: the late afternoon sun, the…
Show 17 more
- In Memoriam, Encountering World War I, and Thinkable Horrors 18 Jul 2026 A delayed education on the trenches and scales of suffering
- One Developer, Two Dozen Agents, Zero Alignment 13 Apr 2026 Why we need collaborative AI engineering and a tour of Ace: the multiplayer coding workspace
- Gas Town’s Agent Patterns, Design Bottlenecks, and Vibecoding at Scale 23 Jan 2026 On agent orchestration patterns, why design and critical thinking are the new bottlenecks, and whether we should let go of looking at code
- January 2026 2 Jan 2026 I entered the new year holding an inconsolable, shrieking baby while London set off an armageddon of fireworks around us. So goes parenthood. The baby is fine, just congested and teething. I am as “fine” as anyone can b…
- A Treatise on AI Chatbots Undermining the Enlightenment 5 Aug 2025 On chatbot sycophancy, passivity, and the case for more intellectually challenging companions
- Vibe Code is Legacy Code 2 Aug 2025 by Steve Krouse
- May 2025 25 May 2025 In a wonderfully dramatic change to my life, I became a mother two months ago. My son was born at the end of March via an unplanned but otherwise uncomplicated c-section. Parenthood has been predictably overwhelming, ex…
- Statistically, When Will My Baby Be Born? 24 Mar 2025 A tiny tool to calculate when your baby might arrive
- ChatGPT Would be a Decent Policy Advisor 13 Mar 2025 Revealed: How the UK tech secretary uses ChatGPT for policy advice by Chris Stokel-Walker for the New Scientist
- March 2025 5 Mar 2025 Well, I've had a dramatic start to the year. Normally, the design agency I joined a short eight months ago, unexpectedly closed down in January. Despite running for a decade and working with almost every major tech comp…
- Humanity's Last Exam 20 Feb 2025 by Center for AI Safety (CAIS) and Scale AI
- DeepSeek 26 Jan 2025 If you're not distressingly embedded in the torrent of AI news on Twixxer like I reluctantly am, you might not know what DeepSeek is yet. Bless you.
- Common Misconceptions in AI 12 Jan 2025 Common Misconceptions About the Complexity in Robotics vs AI by Dan Ogawa
- Undetected AI Exam Answers 11 Jan 2025 A real-world test of artificial intelligence infiltration of a university examinations system: A “Turing Test” case study by Peter Scarfe, Kelly Watcham, Alasdair Clarke, Etienne Roesch
- Unbaited 11 Jan 2025 by Daniel Petho
- Smidgeons 11 Jan 2025 Welcome to the smidgeon stream. This is a new kind of content on the Garden. One that was
- How to Import Academic Papers from Zotero into Tana 7 Jan 2025 How to use Zotero's translator and Tana Paste formatting to easily import papers into Tana
Link verified 20 Sept 2026. Recent items update automatically from the channel.
x.com @Mappletons Posts.
Recent
- The carnage of giving a bunch of nerds a thermal printer to play with 1 Nov 2026
- Backstage at Universe is literal Mission Control 29 Oct 2026
- GitHub Universe day one done! If you’re here, come by tomorrow to chat to all the @GitHubNext folks at the Speakeasy section. We’re in white lab coats. Also demoing some stuff in the final keynote. 28 Oct 2026
Show 17 more
- Mildly performative but very directionally correct. In the sense of having a focused environment to deeply engage with the questions and decisions that matter, and not being drowned in tool calls and other irrelevant implementation details. 26 Oct 2026
- I know I've been prompting my agents correctly when they present interfaces at the grammar, noun, and verb level. Took a lot of skill crafting to get here but god it's refreshing to work at the correct abstraction level and not like "my brother in christ why is this a modal?" 26 Oct 2026
- okay this is actually funny to do though only covers what I’ve said to my personal chatgpt account, not work agents who are separate and have been exposed to all my flavours of stupid 24 Oct 2026
- The trouble with crystallising your thoughts into solid form like this is you spent so much time marinating in the material that when you’re done you’re like “okay this is old news and I have 5 new theories, this is so backwards and passé” (jk obvs good I love it) 24 Oct 2026
- Okay I do have to say riding in Waymos in SF is actually a peak experience. I love these things. Plays your own Spotify music while you cruise around marvelling at the amount of engineering packed into this thing. Wild and beautiful. 23 Oct 2026
- New talk! About the divided human & agent worlds, boundary objects, and building thicker interfaces I've been exploring how to improve "planning", aka. deciding what agents should build, how you want it built, & making the whole team part of that process 22 Oct 2026 maggieappleton.com
- Amazing addition to @GitHubNext 🔥 20 Oct 2026
- It’s time to put your agents down and touch some pumpkin 19 Oct 2026
- A few people asked about my workflow for redoing the agentic workflows website in a day. This is roughly it. Took about 4-5 focused hours + 15 mins over the next few days iterating on messaging with the team behind the project. 8 Oct 2026
- Here's our demo from the https://t.co/06jN5zdbhB hackathon! A self-service agent that helps people deal with "single justice procedures" (small crimes like speeding). Currently done manually by phone, could easily be an always-on agent Mad trailer by @yungbose with @HyperFrames_ 7 Oct 2026
- Found time last week to cleanup the homepage & docs of Agentic Workflows, one of the very successful @GitHubNext projects of the last year. TLDR: run your own software factories on GitHub Actions Previous page was vibed by agents and backend developers. Took me a day to redo it. 7 Oct 2026
- Spent the last two days hacking at Downing St with this wonderful team. Building agentic systems to help solve small inefficiencies in the justice system. Of course just prototypes but still realistic and implementable. Learned a lot. Very cool to be the heart of Westminster. 6 Oct 2026
- Hillclimbing is my new hobby and this is beautiful 29 Sept 2026
- Jigs very good for tweaking design system values with live previews of how those colours work in the context of specific components and WCAG contrast calculators on screen. So easy to say "show me every component that uses grey 400 and put it in the lab file" → tweak → commit 24 Sept 2026
- For people new to design jigs, this is a good example of using them in the wild. I first saw them used by @doctorbaytas in this beautiful talk at @meetgranola: 24 Sept 2026 youtube.com
- The current state of agent design skills 24 Sept 2026
- Had a wonderful chat with @GergelyOrosz - always nice to go deep on ideas in a proper long chat. Got to show off some notebook sketches and design jigs with agents too. 23 Sept 2026
Link verified 20 Sept 2026. Recent items update automatically from the channel.
Mastodon @maggie@indieweb.social Posts.
Link verified 20 Sept 2026.
About the English under a post
Some people here publish in a language other than English. Where they do, this site shows a machine translation beneath the post, in this typeface — the site's own, not theirs.
The post itself is never changed, moved or hidden: what is set in the serif above is exactly what the person published, and it is what to quote them on. A translation can be wrong in ways that matter, especially about tone.
Only the post's own words are translated. A quoted post, a linked article and a belief on korrents.com are left in their original language.