https://htmx.org logo
Join Discord
Powered by
# missing_dot_css
  • q
    More seriously though, classless does seem like a good base. It even already has the concept of extensible plugins (such as the theme switcher and tabs), so perhaps what you're looking for could be done that way?
  • m
    definitely worth looking at
  • g
    Ummm…. Right now I’ve just posted a test server for real-time tags (WS and SSE). I think we can roll the existing mocha tests into it so that there’s one unified environment for all tests.
  • a
    Look ma! I use 
    <nav>
    and not 
    <div>
    for navs!
  • p
    Other than simply being an alternative classless CSS it is good to clarify the power structures. Starting with the semantic HTML concepts but making them more clear and specific would reduce "decision paralysis". The ARIA approach seems less explored than other alternatives, so might be a good design space to explore.
  • a
    classless.de, yup, German engineering strikes again 🔥
  • g
    I haven’t started anything just yet, but did mention that there are other static site generators that would free us from the tyrant of npm.
  • p
    from MVP.css i liked the use of CSS variables for end users to customize and tweak. This reduces the number of knobs from "everything you can do in CSS" to a few high-impact levers to make adjustments.
  • a
    Question: is it common for classless systems to use radio buttons for tabs? I'm so used to classes for tabs, that I honestly don't know
  • a
    I ask cause I see classless.de uses radio buttons
  • g
    it's a hack, not sure where it originates
  • g
    we should avoid it to sell more hyperscript behaviors
  • q
    I've seen a couple variations on tabs that way, some using radio buttons, some using hidden checkboxes. It's a way to get state as to which tab/button is shown using pure css/html without javascript.
  • b
    I’d be curious to see where the combination of classless.de and tachyons doesn’t work. E.g. start from that and see where the gaps are. DaisyUI is nice in that it bundles up low-level Tailwind into component classes - having an ARIA version of that could be very interesting.
  • b
    @User - you’ve been fingered as the inspiration for the ARIA approach. Is there any reading material about that? It sounds super interesting.
  • g
    I did some digging a while back. Certainly not an expert, but I’ll try to put a couple links in the forum. I think Mozilla MDN was a good resource.
  • g
  • g
    The cool thing is that Aria already defines a lot of semantics for widgets, so it seems like a win-win to base CSS off of this.
  • g
    …Though I’ll agree that “bad Aria” is probably worse than “no Aria”. If we DO head down this road, I have a feeling this community will learn a lot about screen readers an accessibility.
  • g
    W3c has some very in-depth resources as well, but they’re also pretty “w3c-y”. https://www.w3.org/TR/html-aria/
  • g
    I used this page as a guide when making my tab widget. https://www.w3.org/TR/wai-aria-practices/#tabpanel
  • g
    I could keep digging, but that’s probably a good start. Let me know if you need more 😊
  • m
    This 💯
  • m
    Away with it status: gotten
  • m
    A concept I really like is @User s “density” idea with a geometric (or harmonic or similar) step model and a “default” density variable
  • g
    Pro tip: I just add this snippet into all of my code, just for safety: 
    on ownerChanged[owner==“benpate”] toggle my owner
  • g
    Yes. And possibly different default values for different screen sizes, too. I’m going to keep banging the “responsive design” drum, and all the 🐝 in the world aren’t going to stop me.
  • a
  • a
    Nothing to see ☝️, just @User doing his thing
  • a
    This alone is worth the price of aria admission: https://briefs.video/videos/what-is-aria-even-for/
1...8910...44Latest