About Bronze - Silver - Gold and scope of views etc.

[spanchang]

Views and processes:
It says a view is akin to a Web page. Sowhile evaluating a process the entire Web page(s) is judged, including common header/footer section, right?
But then there is a feedback question about whether an image in the footer without an alt is critical to the process.
So here is a thought: Grading system: Bronze-Silver-Gold
Bronze: should cover specific views and processes for accomplishing task flows / accessing information.
parts of Web page, like header nav that enable one to find the particular task or section of interest should also be within the scope of this review.
Help documentation linked from the view / process should be within scope
Out of scope: Footer sections / banner / complementary sections as long as they pass non-interference requirements.

Silver: Bronze plus
Cover all declared scope i.e. including footer / banner etc. sections excluded from bronze

Gold: Silver plus
i. Include a mandate to perform usability testing.
ii. Passing selective Level AAA SCs

Note: While above appears to be in the context of Web content, it can be applied to software / apps too.
Thanks,
Sailesh

[Myndex]

Hi @spanchang thank you for commenting.

I am going to respond as to how the conformance model is working right now for Visual Contrast.

All visible text is in scope, except purely decorative (i.e. dingbats of flowers in a line). However, the purpose of a given text element sets different limits and requirements, with further variations per level of conformance.

Brief Overview of Contrasts

In terms of luminance contrast (ignoring size, spatial, hue, and other contrasts) there are four broad categories:

1. Fluent Readability Contrast
2. Contrast for Spot Reading
3. Contrast for non-text objects, images, elements, states (focus), etc.
4. Purely aesthetic Contrasts.

Number 4 has no requirement other than to not interfere with numbers 1 thru 3. Number 1, readability, has by far the highest contrast requirement as needed to enable whole word recognition in the VWFA (Visual Word Form Area) of the brain. Number 2 is a relaxed contrast requirement as spot reading assumes that letter-by-letter lexical processing is acceptable. Number three is a bit more complicated but typically as relaxed as number 2, but in some cases can be more relaxed, yet in other use cases have a higher contrast requirement.

Use Cases Defined

(A) Fluent Readability:

• Columns of body text (body text has elevated requirements for best score)
• Headlines, Subheads, Asides
• Primary navigation, Primary menus
• Descriptive captions that convey meaning beyond the image.
• Font weight and size are interdependent with luminance contrast.

(B) Spot Reading Level

• Copyright, ByLine, and Legal Notices and links thereto (secondary navigation).
• Captions that only restate the captioned item or provide only credits, byline, rights, etc.
• Ancillary text that is not a part of, nor critical to the understanding of, the primary content.

(C) Non-Text Elements

• This means buttons, controls, form fields, clickable objects but does NOT apply to any text within these items.
• This also includes temporal/spatial state changes (including state changes to text, which is necessarily separated from the text's readability contrast).
• Contrasts within an image or drawing that is part of content.
• Contrasts in size, shape, temporal, and spatial relationships are interdependent with luminance contrast.

(UX) Aesthetic Contrast

• Simply anything that is purely decorative, non-readable, not needed for understanding content.
• It can include contrasts that may be helpful for organizing content but are not strictly necessary.
• These have no standard other than to not distract nor interfere with use cases A, B, and C

Conformance Scoring Levels

Preferred

• This is a non-normative "suggested best practices" level. Achieving this achieves Level 4, but does not gain any "bonus" points. It is intended as guidance for more ideal design goals that exceed what is reasonable as a scorable level.
• Includes more restrictive font-size minimums and more restrictive use of low-contrast elements.

Score 4 — "Ideal Minimum"

• The normative minimum for a Score of 4 for the view.
• Case A elements are very similar to the "Preferred" level, but Case B and C elements have more relaxed requirements.

Score 3 — "Acceptable Minimum"

• The normative minimum for a Score of 3 for the view.
• Case A, B, and C elements are only slightly relaxed relative to level 4.
• Intended as a "catch" level for a well designed and well conforming site that "just misses" in one or two areas of visual contrast performance.
• Also, this level is intended to pass sites that are presently passing 1.4.6 contrast enhanced (WCAG 2.x AAA) provided that the background colors are no darker than #CCC and no Case A content uses a font smaller than 16px.

Score 2 — "Marginal"

• The normative minimum for a Score of 2 for the view.
• Case A, B, and C elements are relaxed relative to level 3. In particular the high-contrast Case A is relaxed, and certain minimum sizes of fonts are relaxed.
• Intended as a way for sites that are presently passing 1.4.3 contrast (WCAG 2.x AA) provided that the background colors are no darker than #BBB and no Case A content uses a font smaller than 14px.

Score 1 — "Poor"

• The normative minimum for a Score of 1 for the view.
• Case A, B, and C elements are relaxed relative to level 2, with few restrictions on minimum sizes for Case B text.
• Intended as a catch-all for sites that are presently passing 1.4.3 contrast (WCAG 2.x AA) despite significant design deficiencies and/or false passes of dark color pairs.
• Score 1 means "deficient" but passing, so that sites can still operate and not be penalized (in places where they are by law to comply) to give them time to address and correct their issues.

How This Answers Your Comment

The <header> or <footer> does not denote use cases as defined herein. Nor does the <main> or <article> or... you name it.

It is definitely possible (in fact more than likely) that the header will contain text of the page title, and that is definitely a "Case A use case" and the footer could easily have navigational elements that fall under Case A. Meanwhile, <main> or <section> could easily have Case B and Case C elements.

The point is, that at least for Visual Contrast, use cases do not align with the broader semantic markup of sections and other elements. And definitely NOT header! The closest might be footer, but even then, in the conformance model for Visual Contrast, ALL text is in scope, but there are sub-scope requirements that differ based on use case and score level.

It is the USE CASE that is important, not the semantic markup.

Thank you for posting the comment,

Andy

Andrew Somers
W3 invited Expert
Myndex Color Science Researcher

[davidberman]

I am concerned about both the Anglocentrism (and colorcentrism?) challenges of using the actual terms "Gold", "Silver", and "Bronze" for the levels. Furthermore I’m concerned about unnecessary burden within our industry regarding how we’ll repeatedly express these constructions: I am imagining that professionals will immediately shorten the three tiers to G, S, and B in their documents (in English), and so on. Is this the correct place for me to post this concern, or is it already explored elsewhere?

[patrickhlauke]

I am concerned about both the Anglocentrism (and colorcentrism?) challenges of using the actual terms "Gold", "Silver", and "Bronze" for the levels.

i'd have thought that that categorisation, common throughout international/global sporting events, would be fairly universal

[davidberman]

Thanks for weighing in Patrick; while I agree that the international sport's embracing of the Olympic cliche is universally recognized. However, we are trying to be the best. Relying on colors is not our a11y best... we're almost encouraging people to use non-text indicators in reports. Creating something that won't be instantly recognizable/understandable will be best. We are not giving out medals: we are setting conformance standards that will be cited in regulations as a normative ISO standard. There will be times when "Bronze" will be a regulatory fail rather than "medal-worthy". And sometime "Gold" will be an inappropriate target. And it will still need to be translated. Is this the correct place for me to post this concern, or is it already explored elsewhere?

[mbgower]

At Tuesday's call, the working group had a resolution to remove ratings and scoring in the next charter. When/if that scoring returns in a couple of years, these comments can be brought to consideration again, but for now, I think this issue can be deferred and I have added a label to that effect.

Remove ratings, critical errors, and scorings. Add editor's note similar to 'Critical errors was introduced as a concept in the FPWD. Although it not currently included, AGWG intends to re-asses this concept in the future.'