Eemeli Aro

I don't think this is obsolete. We still ought to say when and what isolation or marks are to be used, as there are multiple options and situations where it's...

Just to make the link explicit, this issue is a blocker for finalising the [bidi-usability](https://github.com/unicode-org/message-format-wg/blob/main/exploration/bidi-usability.md) design doc.

The comment is getting attached to the key of the key-value pair.

They're documented here: https://eemeli.org/yaml/#collections In a parsed document, `key: K` will always be a node of some sort, extending `NodeBase` -- see the preceding section of the docs for that.

A possible solution to the titular question here is presented in https://github.com/unicode-org/message-format-wg/pull/558#issuecomment-1850029898, including a proof of concept parametric XSLT transform.

1. It's quite intentional that the default bidi strategy does not allow for optimizations, as it's meant to produce the same output in different implementations. This enables e.g. rehydration to...

We're not looking to guarantee the same results, but to _enable_ them. If there's a way for a user to get the exact same bidi isolation with two different implementations...

> The expression `{$bar}` is being assigned to `$foo`. This should not be taken to imply that anything _happened_ to `$bar`. Any additional transforms happened to the _expression_ `{$bar}`, not...

> If a user expects that the meaning of `{{{$bar :func} and {$foo}}}` can be determined by replacing `$foo` with `{$bar : func}`, they would be wrong. The spec makes...

The core issue is probably arcane enough that it doesn't really matter, provided that implementations only do reasonable things when resolving `{$bar}`, and functions do similar things to primitive and...