- From: John Hudson via GitHub <sysbot+gh@w3.org>
- Date: Fri, 16 Jul 2021 04:13:01 +0000
- To: public-css-archive@w3.org
I’m not pretending otherwise. I am saying your proposal doesn’t work as a solution. You proposed using different interpreting the opsz scale units differently for web and for print. That isn’t workable because 9pt and 9px are different sizes, so the type designer cannot design something at ‘9’ on the opsz axis scale that is optically tuned for a specific size. A solution _might_ be to register a second axis, so there are dedicated optical size axes for pt sizes in print and px sizes in web. But the problem remains that _px is not a size unit_, so from my perspective this whole issue isn’t about the axis scale definition but about _interpretation_ of the axis scale. Even if you have an axis scale that you define in terms of e.g. the CSS reference pixel—which is as close as this stuff gets to a size unit, so something a type designer could try to target—, you still have interpretation of that scale in rendering to px in various environments that will be only more or less accurate in terms of hitting what the type designer intended. As I said previously, I am okay with that uncertainty, so long as it is at the interpretation level and not something that is forced down onto the level where the type designer is trying to do optical size design. -- GitHub Notification of comment by tiroj Please view or discuss this issue at https://github.com/w3c/csswg-drafts/issues/4430#issuecomment-881162459 using your GitHub account -- Sent via github-notify-ml as configured in https://github.com/w3c/github-notify-ml-config
Received on Friday, 16 July 2021 04:13:03 UTC