Darien Maillet Valentine

@caridy `setTimeout`, `structuredClone`, `atob` etc are operation members of Web IDL global interfaces like Window in HTML. They’re even sensitive to the receiver’s brand and will throw TypeErrors if called...

> As for the various bugs in Chrome (I'm assuming that they are bugs because other browsers do not behave that way) This sounds like [relevant realm](https://html.spec.whatwg.org/#relevant) behavior. I think...

> I think that works because [AllowShared] ArrayBuffer would end up ignored. What makes that ignored? That behavior _does_ make sense to me, but I don’t think it’s what’s currently...

@annevk great idea — that’s a neater solution all around for the cases I described. new issue opened: #762

Looks like this typo made it into [Web Platform Tests](https://github.com/web-platform-tests/wpt/blob/7b0ebaccc62b566a1965396e5be7bb2bc06f841f/mimesniff/mime-types/resources/mime-groups.json#L11-L12) too.

This seems akin to how [DOMException interface objects inherit from %Error%](https://webidl.spec.whatwg.org/#es-DOMException-specialness). I don’t see how defining an AbstractModuleSource Web IDL interface would help with it — that would create a...

Declarative comment / prose text “overrides” both have an unclear relationship to the imperative realm-initializing algorithms in Web IDL (_when_ do the “overrides” happen?) and make it impossible for tooling...

In addition to what @saschanaz said, new Web IDL syntax probably wouldn’t make sense here because this is really about the ES binding (which is what the extended attributes defined...

@domenic It does look like that — two implementors stated a preference for prose, and I’m not an implementor (at least not in a sense that I would expect to...

How does is-callable currently handle platform classes, which already return the generic native code string?  Or bound constructors?  If NativeClass were added, I would hope it would also...