Hacker Newsnew | past | comments | ask | show | jobs | submit | waldir's commentslogin


Correction: Only Jacob Thornton is one of the co-founders of Pierre; the other one is Ian Ownbey (sources: https://presentofai.com/spotlight/pierre.co, https://cellcog.ai/blog/openai-acquires-pierre/)


Do you have a comparison page, or a list of the annoyances you fixed from PicoCSS, accessible anywhere?


For some reason my detailed reply got flagged...

So if you can't see it: feature-comparison matrix plus notes from migrating several real projects from Pico:

https://github.com/Leftium/nimble.css/blob/main/specs/pico-m...


Oh, indeed I can't see any reply other than this one :(

Thanks for the link — that's pretty comprehensive!


Apparently the other reply got auto-flagged because it was detected as genai. Let me paraphrase that comment (easier than reading that entire MD document, which was AI-written):

The main annoyances:

- oversized buttons

- 100+ root custom properties in devtools

- fighting against too many breakpoints (unpredictable font/container sizing)

---

I also changed/added some features:

- much smaller core: only 4.1 KB Brotli

- CSS cascade layers so app CSS can override more easily

- a surface color system (a la Open Props)

- OKLCH colors with auto derived hover/focus states and light-dark() theming

- centered layout with real gutters, plus content bleed outside gutters

- .no-nimble opt-out class (instead of .pico opt-in)

- ability to configure universal control-density (buttons, inputs, selects)


The main annoyances:

- oversized buttons

- 100+ root custom properties cluttering DevTools

- fighting against breakpoint-heavy font and container sizing

- container "padding" that could disappear at exact breakpoints

- having to fight framework styles with specificity

Nimble uses more compact controls, ~20 semantic variables, simpler grid-based centering with real gutters, and cascade layers so normal app CSS wins.

---

I also added some features/changed:

- much smaller core: Nimble is only 4.1 KB Brotli

- cascade layers, so normal app CSS can override the library without specificity fights

- a surface hierarchy for page/card/input/overlay backgrounds (a la Open Props)

- OKLCH colors with derived hover/focus states and light-dark() theming

- centered layout with real gutters, plus bleed-edge, bleed-wide, and bleed-full

- .no-nimble to opt third-party component subtrees out of Nimble styling

- shared control-density customization for buttons, inputs, selects, etc.

- .nc-join for visually joined controls without coupling that styling to ARIA role="group" (this kept confusing coding agents; technically they should not be coupled)

---

Feature-comparison matrix plus notes from migrating several real projects from Pico: https://github.com/Leftium/nimble.css/blob/main/specs/pico-m...


Were you aware of git-bug when you created gitsocial? If so, which patterns did you reuse and which ones did you decide to diverge from? Or would you say the two projects embody fundamentally different visions?


No, I was not! GitSocial originally, in Jan 2025, started as a Twitter/Bluesky over git app (as a VSCode extension with just social features), and then pivoted to being a full-fledged git platform, to include issues, PRs, releases, etc. You can still "follow" repositories though!


Got it, thanks for the context!


> It's the difference between instructions for a user and instructions for a developer

Sounds like a job for a CONTRIBUTING.md :)


They've already started a public repository here: https://github.com/anthropics/skills (It's the last link in the "Getting started" section of the post.)


I get why people may not like the comparison you're making, but that aside, I agree that the article insists uncomfortably often that the freedom only comes if "the work" gets done (in various phrasings). As if to ensure that nobody gets the impression that a pleasant work experience is something anyone deserves rather than a generous gift by such a magnanimous employer.


Extensive discussion on reddit, dating back from the day after the initial announcement date (15 February): https://www.reddit.com/r/github/comments/113dm5v/anyone_get_...


So... was your project one of the 20 accepted ones? I got the email but my project (tldr-pages) wasn't selected.


Doesn't that suggest that tooling wasn't the issue after all?


The lack of tooling for XSLT and modern JavaScript stuff are on different sides of the adoption curve. XSLT lacked tooling on the upside of the curve so it never hit critical mass. Its other problems could have been solved had it hit critical mass and seen wider adoption.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: