Is there an intended minimum viewport size for 1.4.4?

[daniel-montalvo]

Ref:
act-rules/act-rules.github.io#1318

For rule:
Zoomed text node is not clipped with CSS overflow (59br37)

The rule currently makes the following assumption:

While success criterion 1.4.4 Resize text does not explicitly mention which viewport size has to be resized up to 200%, it is assumed that a viewport size of 1280 by 1024 is applicable. A 1280 by 1024 viewport size is explicitly mentioned under success criterion 1.4.10 Reflow.

Is this assumption reasonable even though 1.4.4 does not mention any viewport size?

[mraccess77]

While 1.4.4 doesn't specifically provide a value - I'd treat it like an accessibility supported issue and what you are making the claim against. 1280 is a width to start from and is what I use for testing that.

[patrickhlauke]

we've had a lengthy back-and-forth previously on this topic - see #704

no hard decision has been made as far as I'm aware. I still contend that, by the way the SC is written, there is theoretically no minimum viewport size. while yes, there are complications that then arise with responsive web design, and how to deal with font sizes that change when viewport size changes, the concept that "at whatever viewport size the user is, they should be able to somehow make the text 200% bigger at least" still holds for me.

[mraccess77]

200% at 640 is 320 -so that does align with our SC 1.4.10 - however, SC 1.4.10 doesn't focus on text actually becoming any bigger (although it is assumed that in most cases it will - it sometimes does not).

[patrickhlauke]

1.4.4 also doesn't care if there's bidirectional scrollbars as a result or not, so the connection with 1.4.10 is very loose at best.

but again noting that if you essentially suggest that 640 CSS px width viewport is the "minimum" it means that any "mobile"-sized site that somehow blocks text resizing (including zooming) would be passing 1.4.4, which...wouldn't be good.

[alastc]

We have a few open issues in this area. (Chair hat off for the rest of this comment.)

I think what the ACT rule outlines is a good practical approach to testing, but when it comes to mobile we have a few nuances:

• For web on mobile, pinch-zoom (if not disabled) could be seen as passing, as there is no restriction on bi-directional scrolling.
• For those applying WCAG to native, there are some impossible scenarios to overcome, e.g. Android restricting text-sizing to ~130%. See Looking for guidance regarding SC 1.4.4, Resize Text, on mobile native apps wcag2ict#4, for which an update to WCAG2ICT would be required.

Patrick does have a point, we just need a little time to package these issues and get the group's attention on it (after we've worked through the WCAG 2.2 issues).

[mraccess77]

On mobile I agree pinch zoom would meet. While Android's large text feature could be used - it is limited but some browsers on Android likely go higher with there own settings.