What I understand of this:
- There was a PR from an open-source contributor
- He checked out the branch.
- In the branch, there might have been a new npm dependency that ran a post-install script, which exfiltrated his npm credentials.
See www.mediawiki.org/wiki/Page_Pr...
The 4 principles are respected.
I don't understand.
Do you have reference material from WCAG that would support this? As far as I know, this component respects www.w3.org/WAI/WCAG21/U....
As I understand it, a screenreader user can't know if a link has a preview card or doesn't have one, it's invisible. The experience for them is the same.
So if they want to navigate to see the content of the link, they navigate, if they don't want to, they don't.
It's a preview of the link, so the content is accessible to user who rely on assistive technology.
I don't understand what's not accessible about this React component?
This is to show a preview of the link that will be clicked on. It's designed to be invisible for VoiceOver, JAWS, or NVDA users.
The WCAG 2.1 success criteria seem OK, e.g.
2.1.1 Keyboard: "Enter" shows the same content.
We are literally trying to find a solution to this on the Material UI button loading state right now. A proper fix would be amazing.
Switch.Root.Props
We could do more to ease the switch, e.g. github.com/mui/base-ui/..., but unclear if it's needed.
The APIs were design by starting from the Radix Primitives API, and then evolving them based on:
- Learnings from Radix Primitives
- Learnings from Material UI test suite
- React 19 new API, e.g. now the ref is a regular prop
Everyone is asking about the differences between Radix and Base UI, which is an upcoming headless UI library we’ve been working on.
@base-ui.bsky.social is here
Watch out for news. I’ve been working on the Base UI docs for the past few months, and this is going to be the component library you didn’t know you needed.
Yay. @base-ui.com just went live 🎉
It looks super promising. I've been eagerly anticipating this because, on paper, it has the potential to combine the best of the Radix API with the ambitions of MUI components and their use cases.
It's too easy to forget that end-users have slow hardware too.
I have reality appreciated working with slow hardware in the past. Any performance regressions becomes so obvious.
Sometimes the best engineering happens when you’re just tinkering, free from expectations and deadlines.
