jfkthame

> What does Microsoft and Apple rasterizer do with such fonts? We wouldn’t usually allow fonts they can’t handle, and specially if they don’t fail gracefully. cc @jfkthame Yeah, that's...

Windows 10 refuses to install the font; it tells me it's "not a valid font file". I was able to load it as a webfont resource in Firefox after hacking...

Windows still refuses to install the font, insisting that it is "not a valid font file". So while we'd be able to use such a file as a webfont (with...

Yes, I wondered if something like that might be the answer. So in browsers that can handle the font, we could configure it to allow this, but a standalone OTS...

I think I'd agree that `font-synthesis-style: none` should disable fallback from italic to oblique, whether "real" or synthetic. It should also disable synthesis of oblique angles outside the available range...

> There are some ZWJ emoji sequences that do not have VS, compare https://www.unicode.org/Public/emoji/latest/emoji-zwj-sequences.txt. Should `font-variant-emoji` property affect any of the codepoints in these sequences? Yes, in my opinion: the...

> Firefox (and Chrome)'s current behavior is better than the specced behavior, so I think we should update the definition of `auto`. Given their current behavior, it's presumably web-compatible to...

> I have a question about `text-emphasis-position: auto;` I assume that `auto` is the initial value, is my assumption here correct? I believe that was the intention, yes.

Yeah, I'm not convinced we'd get much value out of this. In principle I can imagine using such an API to optimize some style-change handling, but I wonder how often...

> For `ic`, `1em` is the correct size for the vast majority of CJK fonts, so it might still be tempting to synthesize the missing measurement to that value, but...