Re: [w3ctag/design-reviews] Question: should `shadowrootadoptedstylesheets` perform a fetch? (Issue #1195)

jyasskin left a comment (w3ctag/design-reviews#1195)

Hello! Sorry that it took us a long time to get back to you.

We do have consensus that the `shadowrootadoptedstylesheet` attribute should fetch, even if that risks a FOUC. We also _don't_ think that the [`host-for` attribute](https://github.com/w3ctag/design-reviews/issues/1195#issuecomment-4361332377) would be an improvement.

Ideally, we'd like to see measurements before being confident that adding [thousands of `<link>` elements](https://github.com/w3ctag/design-reviews/issues/1195#issuecomment-4362055519) is a significant expense compared to thousands of stylesheet attachments using an attribute. However, it's a plausible claim even without the measurement, so we don't object to using the `shadowrootadoptedstylesheets` attribute.

Harking back to https://github.com/w3ctag/design-reviews/issues/1000#issuecomment-3254588565, it would be nice if the `shadowrootadoptedstylesheet` attribute were syntactic sugar for some `<link>` element so that a developer who needs more control than the attribute provides (for example to control render-blocking and error-handling), could smoothly transition to the more expressive element. For example:

```html
<template shadowrootmode="...">
  <link rel="adoptedstylesheet" specifier="https://example.com/styles" blocking="render" onerror="handleError()">
</template>
```

However, we recognize that a large fraction of the feature's benefit is given by the attribute, and that Javascript fallbacks are already possible, so we don't want to block the overall feature on concurrently finding and standardizing the ideal `<link>` design. Please do consider this as a follow-up, especially if developers ask for more control.

Thanks to @dandclark for drafting several versions of this comment and dealing with my slow replies.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/w3ctag/design-reviews/issues/1195#issuecomment-4608033622
You are receiving this because you are subscribed to this thread.

Message ID: <w3ctag/design-reviews/issues/1195/4608033622@github.com>