Re: [editing-explainer] replaceText input event could span multiple Typing Nodes (#50)

> The browser would probably guess that "word" is a single word.

No, the browser would probably take into account the style applied to the Element. In fact I ran the exact code you offer above in Safari and it correctly identifies that anything in the span is a different word given the CSS provided. I don't think we'd have to specify when things are words and when they are not. The UAs are capable (and should) do it themselves if they want to offer spell-check across boundaries.

Note that "real" text editors offer spell-check of partially-styled words. It follows then that "proper" web-based editors should eventually do the same. But if the spec doesn't allow for it natively, then it's the editor devs who are going to have to write reams of JS to get around that problem.

I think it's a very bad idea to bake into the spec a limitation such as this one. Specs live long lives and defeating future capabilities for the sake of short term simplicity is something I can't get behind.

---
Reply to this email directly or view it on GitHub:
https://github.com/w3c/editing-explainer/issues/50#issuecomment-105300736