Issue 78 SC text (was Re: Post-Minutes update)

Hi Alastair,

Thank you very much for your email and reading yesterday's minutes. We
missed you at the meeting.

What would you think the issue 78 SC text should be? On GitHub it currently
reads:

No loss of content or functionality on a webpage is caused by overriding:

1. font family to Verdana, or

2. foreground and background to a single different foreground color and a
single different background color, or

3.  line-height of all text to 1.5, letter-spacing to 0.12em, and
word-spacing to 0.16em.

Note: If no mechanism exists to override these items on any user agent for
the target technology, then the author has no responsibility to create one.

Thank you.

Kindest regards,
Laura
On Feb 9, 2017 7:25 PM, "Alastair Campbell" <acampbell@nomensa.com> wrote:

> Hi everyone,
>
> I'm sorry I couldn't be there, but when reviewing the minutes we seem to
> be going back over old ground again? I'll try and keep this short...
>
> I'm not aware of much push back against the need to avoid horizontal
> scrolling, regarding Resize and Linearise:
> - The only issue left on Resize content is how to deal with IDE type
> interfaces, so I'm working on an exception or subtle language change for
> that.
> - There is some push back on the mechanism of linearisation, not the need.
> I.e. if a user over-rides the layout, what does the author need to do? That
> is a good question, I've outlined that we need to get the SC into the
> draft, and then work on testing that leads to techniques & failures.
>
> Regarding the "ability to override", everyone needs to understand this:
>
> A USER REQUIREMENT IS NOT A CONTENT REQUIREMENT.
>
> (Sorry for shouting, I'm in a plain text interface, no bold available.)
>
> So if the SC states "The user can override X", that is always true because
> the user CAN override it. Does that mean it won't break and become
> unusable? No.
> That approach does not work in WCAG 2.x.
>
> It has to go the next step, and say: if you do X to the content, it is
> still usable & readable.
>
> And in this case X has to be specified, which is where Verdana, the EM
> sizes and the black & white came in. That doesn't mean that LV people will
> use those values, but the issues *triggered* by over-riding
> font/colour/spacing will be triggered by that test.
>
> The second issue is how much things like spacing can be pushed. I'm sure
> anyone who has tried overriding the spacing has run into issues where
> things overlap or layouts fall apart.
>
> If the values for spacing are greater than what can be accommodated in the
> 'normal' layout of most sites, it moves into the realm of personalisation,
> or linearisation where you completely override the layout anyway.
>
> So a conversation about the 'best' or 'largest' word/letter spacing values
> is irrelevant, because either it fits within regular layouts (which need to
> allow for *some* variation because of different display of fonts,
> internationalisation and other factors), or you deal with it as a
> personalisation issue, in which case it is not a content requirement.
>
> This approach of enabling the user to override the presentation (and
> requiring testing of it) is new for WCAG, and there is a lot of work to do
> to prove that it is feasible, see this long thread on that approach and the
> push back: https://github.com/w3c/wcag21/pull/89 (That's linearise, but
> the same applies to override.)
>
> If an SC requires personalisation, I have my doubts that it will get into
> 2.1. Maybe 2.2 or silver, but it needs the user-agent end to be ready in
> the same timeline.
>
> That is why I'd rather work on the override approach, because it can work
> sooner, and the work to test it will be a good building block for
> personalisation later anyway.
>
> Cheers,
>
> -Alastair
>