I'm a web developer securing the world. I build things for the internet. HTML5,
28 stories
·
3 followers

Quoting Jasmine Sun

2 Shares

If you tell a friend they can now instantly create any app, they’ll probably say “Cool! Now I need to think of an idea.” Then they will forget about it, and never build a thing. The problem is not that your friend is horribly uncreative. It’s that most people’s problems are not software-shaped, and most won’t notice even when they are. [...]

Programmers are trained to see everything as a software-shaped problem: if you do a task three times, you should probably automate it with a script. Rename every IMG_*.jpg file from the last week to hawaii2025_*.jpg, they tell their terminal, while the rest of us painfully click and copy-paste. We are blind to the solutions we were never taught to see, asking for faster horses and never dreaming of cars.

Jasmine Sun

Tags: vibe-coding, coding-agents, claude-code, generative-ai, ai, llms

Read the whole story
yayadrian
9 days ago
reply
Leicester, UK
Share this story
Delete

I thought this said “Winamp” and I got really excited....

1 Comment
I thought this said “Winamp” and I got really excited.
Read the whole story
yayadrian
10 days ago
reply
So close, yet so far
Leicester, UK
Share this story
Delete

Quoting Liz Fong-Jones

2 Shares

In essence a language model changes you from a programmer who writes lines of code, to a programmer that manages the context the model has access to, prunes irrelevant things, adds useful material to context, and writes detailed specifications. If that doesn't sound fun to you, you won't enjoy it.

Think about it as if it is a junior developer that has read every textbook in the world but has 0 practical experience with your specific codebase, and is prone to forgetting anything but the most recent hour of things you've told it. What do you want to tell that intern to help them progress?

Eg you might put sticky notes on their desk to remind them of where your style guide lives, what the API documentation is for the APIs you use, some checklists of what is done and what is left to do, etc.

But the intern gets confused easily if it keeps accumulating sticky notes and there are now 100 sticky notes, so you have to periodically clear out irrelevant stickies and replace them with new stickies.

Liz Fong-Jones, thread on Bluesky

Tags: bluesky, ai-assisted-programming, generative-ai, ai, llms, context-engineering

Read the whole story
yayadrian
35 days ago
reply
Leicester, UK
Share this story
Delete

Ira Glass’s Subway Take is genuinely shocking: “Every podcast is better at...

2 Comments
Ira Glass’s Subway Take is genuinely shocking: “Every podcast is better at 2.0 speed!”

💬 Join the discussion on kottke.org

Read the whole story
yayadrian
105 days ago
reply
I knew it!
Now it’s official, Ira Glass is clearly right
Leicester, UK
Share this story
Delete
1 public comment
jheiss
100 days ago
reply
I nearly always watch YouTube at 1.5x or 1.75x. 2x is usually a little too fast for me. (Hint for YouTube, < and > let you adjust the speed from the keyboard.)

The AI water issue is fake

2 Shares

The AI water issue is fake

Andy Masley (previously):

All U.S. data centers (which mostly support the internet, not AI) used 200--250 million gallons of freshwater daily in 2023. The U.S. consumes approximately 132 billion gallons of freshwater daily. The U.S. circulates a lot more water day to day, but to be extra conservative I'll stick to this measure of its consumptive use, see here for a breakdown of how the U.S. uses water. So data centers in the U.S. consumed approximately 0.2% of the nation's freshwater in 2023. [...]

The average American’s consumptive lifestyle freshwater footprint is 422 gallons per day. This means that in 2023, AI data centers used as much water as the lifestyles of 25,000 Americans, 0.007% of the population. By 2030, they might use as much as the lifestyles of 250,000 Americans, 0.07% of the population.

Andy also points out that manufacturing a t-shirt uses the same amount of water as 1,300,000 prompts.

See also this TikTok by MyLifeIsAnRPG, who points out that the beef industry and fashion and textiles industries use an order of magnitude more water (~90x upwards) than data centers used for AI.

Tags: ai, ai-ethics, ai-energy-usage

Read the whole story
yayadrian
108 days ago
reply
Leicester, UK
Share this story
Delete

I like AI slop and I cannot lie

1 Comment

I looked in my home directory in my desktop Mac, which I don’t very often (I run a tidy operation here), and I found a file I didn’t recognise called out.html.

Here is out.html.

For the benefit of the tape: it is a half-baked GeoCities-style homepage complete with favourite poems, broken characters, and a "This page is best viewed with Netscape Navigator 4.0 or higher!" message in the footer.

The creation date of the file is March of this year.

I don’t know how it got there.

Maybe my computer is haunted?


I have a vague memory of trying out local large language models for HTML generation, probably using the llm command-line tool.

out.html is pretty clearly made with AI (the HTML comments, if you View Source, are all very LLM-voice).

But it’s… bad. ChatGPT or Claude in 2025 would never make a fake GeoCities page this bad.

So what I suspect has happened is that I downloaded a model to run on my desktop Mac, prompted it to save its output into my home directory (lazily), then because the model was local it was really slow… then got distracted and forgot about it while it whirred away in a window in the background, only finding the output 6 months down the line.


UPDATE. This is exactly what happened! I just realised I can search my command history and here is what I typed:

llm -m gemma3:27b ‘Build a single page HTML+CSS+JavaScript UI which looks like an old school GeoCities page with poetry and fave books/celebs, and tons and tons of content. Use HTML+CSS really imaginatively because we do not have images. Respond with only the HTML so it can be run immediately’ > out.html

And that will have taken a whole bunch of time so I must have tabbed elsewhere and not even looked at the result.


Because I had forgotten all about it, it was as if I had discovered a file made by someone else. Other footprints on the deserted beach.

I love it.

I try to remain sensitive to New Feelings.

e.g…

The sense of height and scale in VR is a New Feeling: "What do we do now the gamut of interaction can include vertigo and awe? It’s like suddenly being given an extra colour."

And voice: way back I was asked to nominate for Designs of the Year 2016 and one my nominations was Amazon Echo – it was new! Here’s part of my nomination statement:

we’re now moving into a Post PC world: Our photos, social networks, and taxi services live not in physical devices but in the cloud. Computing surrounds us. But how will we interact with it?

So the New Feeling wasn’t voice per se, but that the location of computing/the internet had transitioned from being contained to containing us, and that felt new.

(That year I also nominated Moth Generator and Unmade, both detailed in dezeen.)

I got a New Feeling when I found out.html just now.


Stumbling across littered AI slop, randomly in my workspace!

I love it, I love it.

It’s like having a cat that leaves dead birds in the hall.

Going from living in a house in which nothing changes when nobody is in the house to a house which has a cat and you might walk back into… anything… is going from 0 to 1 with “aliveness.” It’s not much but it’s different.

Suddenly my computer feels more… inhabited??… haunted maybe, but in a good way.


Three references about computers being inhabited:

  1. Every page on my blog has multiplayer cursors and cursor chat because every webpage deserves to be a place (2024) – and once you realise that a webpage can show passers-by then all other webpages feel obstinately lonely.
  2. Little Computer People (1985), the Commodore 64 game that revealed that your computer was really a tiny inhabited house, and I was obsessed at the time. LCP has been excellently written up by Jay Springett (2024).
  3. I wrote about Gordon Brander’s concept for Geists (2022). Geists are/were little bots that meander over your notes directory, "finding connections between notes, remixing notes, issuing oracular provocations and gnomic utterances."

And let’s not forget Steve Jobs’ unrealised vision for Mister Macintosh: "a mysterious little man who lives inside each Macintosh. He pops up every once in a while, when you least expect it, and then winks at you and disappears again."


After encountering out.html I realise that I have an Old Feeling which is totally unrecognised, and the old feeling, which has always been there it turns out, is that being in my personal computer is lonely.

I would love a little geist that runs a local LLM and wanders around my filesystem at night, perpetually out of sight.

I would know its presence only by the slop it left behind, slop as ectoplasm from where the ghost has been,

a collage of smiles cut out of photos from 2013 and dropped in a mysterious jpg,

some doggerel inspired by a note left in a text file in a rarely-visited dusty folder,

if I hit Back one to many times in my web browser it should start hallucinating whole new internets that have never been.


More posts tagged: ghosts (7).

Auto-detected kinda similar posts:

Read the whole story
yayadrian
130 days ago
reply
You had me a Geocities
Leicester, UK
Share this story
Delete
Next Page of Stories