Open svgeesus opened 2 years ago
Hi Chris @svgeesus
In fact, there are color (hue/chroma) modules that extend the functionality of APCA. There are a variety or reasons that they are not being presented at this time, I'd be happy to discuss via email.
Nevertheless, all recent research does show that as far as readability, it's all about the luminance channel, and chroma/hue is not additive to luminance contrast perception (Legge — in that study, fairly low spatial frequency stimuli were used.)
This particular page (the first getting started page in the HowTo tab) came out of the summer 2019 Visual Contrast group, and as there were a lot of other pressing issues before the FPWD, never had a re-write... In re-reading it, that's probably a good idea do now. Part of the struggle at the time was reducing things to plain language, or explaining with minimal jargon. I feel that I have a better handle on that today,
Literally everything else I've written in more recent years has clearly stated that the "Big Kahunas" of readability are spatial frequency and luminance contrast.
The recent writing and developmental work has not been added into Silver, but from your recent comments I'm working on that task now.
Thank you for reading,
Andy
Hi Chris @svgeesus
To Add: Actually, I did already rewrite that paragraph in PR #630 but it has yet to be merged, and I don't know when it will be. I've been considering setting up a repo with a version that I can keep up to date to be viewable...
On Visual contrast of text, the "Getting Started" tab, I currently read:
This is a good description of total contrast, but I just wanted to point out that neither WCAG 2.1 contrast nor APCA contrast take into account chromatic contrast (the contrast caused purely by hue differences, when the luminance of text and background is identical). Instead, WCAG 2.1 solely considers luminance contrast and APCA estimates lightness contrast.
Possible ways forward:
a) Modify the introductory text, making it clear such as
or
b) Add a way to measure chromatic contrast and a way to combine this with lightness contrast to obtain a total score for visual contrast.
I suggest that option a) would be the simpler and faster one to implement.