1000 results for “I_Hardly_Know_Don”
I had an AI write an Agent Skill for authoring RSSHub routes
Not much of a coder, but I always wanted to contribute. Agents made it practical.
So I had an agent mine 2,727 route PRs and 18,680 review comments, and distilled them into a skill: write routes that survive review, not just routes that run.
🔸 Grounded in AGENTS.md and the official docs
🔸 Opt-in feedback loop
🔸 AI-drafted, unaffiliated with RSSHub. AI errs; scrutiny welcome.
I setup a small photo booth for the minis…
I mean, I could've told 'em that for zero tokens...
I miss the old internet. This version of everything really makes me use it less and less. When even a favorite blog makes you run the gauntlet to verify that I'm human it makes me want to throw in the towel somedays.
I really know nothing about the folks on the committee aside from some remarkably long tenures, but if the medium is the message, LinkedIn is... not great
... I was having a moment, sorry 'bout that.
I started a list of links about designing for VR back in 2015. The domain just turned 10, so The UX of VR has had a proper refresh.
176 resources now, with a short “start here” path and new sections on visionOS, Android XR, WebXR, comfort and locomotion, input and accessibility. Dead links point to the Wayback Machine.
https://www.uxofvr.com #VR #XR #UXDesign #WebXR #SpatialComputing
I wrote this haiku while dropping the kids off at the pool this morning:
The weary pony
Preying on the brewery
Next to a dentist
Be grateful, for #JohnMastodon has spoken to you.
I was certain of 1
🧠 Quiz of the Day - Oct 25
🟩 🟩 🟩 ⬜ 🟩
🟩 ⬜ ⬜ ⬜ ⬜
Score: 5/10
Think you can do better? quizoftheday.co.uk
I thnk it is important to know "the
10 popular misconceptions about Ukrainian history". I learned a lot listening to this Kyiv indepependent youtube at
https://youtu.be/6p4CWUtAxls
I used to treat ChatGPT like a vending machine. Need a React hook? Drop a coin. Need a PR summary? Drop another. Every ask started from zero — stack, conventions, what “done” meant — and every reply drifted a little from how we actually ship.
The waste wasn’t the tokens. It was re-explaining our frontend rules every single time, then cleaning up the same mistakes.
So I stopped pasting one-offs and started treating prompts like code: named, versioned, and checked into the repo.
One-off asks don’t compound
A throwaway chat is fine for a throwaway question. Frontend work rarely is throwaway:
You get something that compiles. You don’t get something that matches the last three PRs. The next person (or the next you) pays the tax again.
What belongs in a prompt library
I keep a small folder — prompts/frontend/ — not a novel. Each file is one job, one outcome shape.
Mine look like this:
prompts/frontend/
pr-review-ai.md # what the model checks vs leaves for humans
new-component.md # scaffold + a11y buckets + token rules
design-token-change.md # changelog / RFC checklist before merge
test-plan.md # real coverage vs fake green
agent-ui-diff.md # component boundaries, no random token edits
accessible-react.md # roles / keyboard / contrast report required
If a prompt isn’t reused at least twice a month, it doesn’t earn a file. Sticky notes in chat stay sticky notes.
The shape every prompt gets
Same skeleton so humans and agents know where to look:
# <job name>
## When to use
One sentence. If it doesn’t match, don’t force it.
## Context to paste
- Framework / router / form lib (or “read from repo”)
- Design tokens source of truth
- Related golden files (paths)
## Hard rules
Bullet list the model must not violate.
Prefer native HTML, keep tokens, no drive-by refactors, etc.
## Output format
Numbered sections the reply must include.
Missing a section = reject the turn.
## Done looks like
Concrete acceptance — not “looks good.”
The “output format” line is the whole game. Models fill schemas. They invent vibes when you don’t give one.
From ChatGPT paste to agent / editor default
Once the files exist, wire them where work actually happens:
prompts/frontend/ or inline the two you use daily.pr-review-ai.md as the system brief; humans still own product judgment.@new-component expands the scaffold prompt so nobody retypes it.You don’t need a fancy prompt CMS. Markdown in git is enough. Review prompt changes in PRs the same way you’d review eslint config.
A tiny migration story
Week 1 I had twelve ChatGPT threads titled “modal again.” Week 3 I had six files and deleted half the threads.
What changed in practice:
Same models. Less re-prompting. The library held the team memory the chat UI kept forgetting.
What not to put in the library
A prompt library is DX infrastructure. If it reads like a blog draft, move it to the blog.
Start with three, not thirty
Tonight, pick the three asks you keep repeating. Write each as one markdown file with When / Rules / Output / Done. Drop them in the repo. Point your editor agent at that folder.
Next time you open a blank ChatGPT box for a frontend task, open the matching file instead. Paste once. Reuse forever. That’s the whole upgrade — from one-off asks to a library your future self (and your agents) can actually find.#frontend #ai #dx #productivity #software #coding #development #engineering #inclusive #community
From One-Off ChatGPT Asks to a Reusable Frontend Prompt Library
I could use some help this week. My phone is about to be disconnected and I'm trying to keep service on until a freelance payment clears at the end of the month. It's $33.66. I'm also out of the sodas and drink mixes that make things here feel a bit more bearable. If you can spare anything, one-time tips help cover both: ko-fi.com/rarebird15 and paypal.me/LanieM. Thank you to everyone who's been able to help.
#Disability #MutualAid
I was wondering when Laura Malvoyante would be getting her new Jolla Phone 2026. Apparently she now has had it for a week and she has published a video on her YouTube channel with her first impressions. And what a video it is! She has been daily driving the Jolla C2 phone and now switched to the Jolla J2.
I'm pretty sure she will follow this video up with one or two more videos, so I recommend subscribing to her channel.
I know VPNs exist, but if there was an option in the Mastodon admin interface to only allow new users to register if they have an IP-address that resolves to Norway... that would get rid of 99% of spam sign-ups and I would probably open registration (as opposed to requiring approval).
#Mastodon #MastoAdmin #Administrators #Allheimen #Fediverse #Norway #Norge #MastodonFeedback
I wonder how many Claude videos on YouTube I need to click "Not interested" on, before the algorithm gets the hint?
I have never read a poem of my favorite poet. Wisława Szymborska (https://en.wikipedia.org/wiki/Wis%C5%82awa_Szymborska) wrote in Polish, a language I cannot read. But there are excellent translations of her work into German (Karl Dedecius) and English (Clare Cavanagh and Stanisław Barańczak). They are excellent in the sense of the translated poems being excellent. I am neither able to judge the quality of the Polish originals, nor the accuracy of the translation but I have to trust the reputation of Szymborska and the ability of the translators.
If AI can generate poetry that is ranked higher by humans than the products of human poets (https://www.theguardian.com/books/2024/nov/18/ai-poetry-rated-better-than-poems-written-by-humans-study-shows), maybe AI it is also better at translating? I have made a small test to get an idea. I took the last stanza of a famous poem ("Some People Like Poetry") by Szymborska, ran it through a widely used current LLM, and compared the outcome to a human translation. My test is of course inadequate to answer the above question, I should have done a blinded study with more poems, more poets, more readers.
The original:
"Poezję –
tylko co to takiego poezja.
Niejedna chwiejna odpowiedź
na to pytanie już padła.
A ja nie wiem i nie wiem i trzymam się tego
jak zbawiennej poręczy."
From https://www.szymborska.org.pl/szymborska/wiersze/niektorzy-lubia-poezje/
In the next post of this thread I show you the two translations, the human one and the one by the LLM.
(part 1 of 2)
I may have reinvented barbecue sauce from first principles.
I made a decent start on my second test model for my Dark Angels 5th Battle Company.
#MiniaturePainting #Miniatures #Wargaming #40k #Warhammer40k #DarkAngels #SpaceMarines #5thBattleCompany #ProperSpaceMarines #PrePrimarisHeresy
I am not your rolling wheels
I am the highway
I am not your carpet ride
I am the sky
I am not your blowing wind
I am the lightning
I am not your autumn moon
I am the night, the night
(Audioslave - I Am The Highway)
Camera: Olympus PEN F
Edit: cropped, otherwise SOOC
#photo #photography #autumn #leaf #SilentSunday #blackandwhite #monochrome #ground #fallingLeafs #beforeWInter #SommerIsOver
I'm making a seasonal anime guide app, in the open for all to experience and learn from.
Let’s finish off the rebase cleanup!
#Jiiiii #DevStream #tvOS #visionOS #macOS #iOS #iPadOS #Anime #SwiftLang #SwiftUI #Vapor #WebAuthn #BuildInPublic #TestFlight #PWA #WebPush
Come chill with me: https://youtube.com/live/XxpLeA-WWEo
I wonder which took longer? Counting 28 teeth on a sprocket in a photo or using the LLM that makes us so efficient and more productive?
I know which approach burned the most non-renewable energy sources.
I wrote this haiku many years ago, whilst on my gap year in Palikir:
The statistician
Noisier than an eyebrow
Next to the toothbrush
#JohnMastodon has cast those pearls before you. Do with them what you will.
I keep seeing Instagram posts from the Financial Times. These days it styles itself as “THE FT” and I swear the space between the E and the F gets smaller every day.
