W3C

Results of Questionnaire Silver Editor's Draft Update and Publish

The results of this questionnaire are available to anybody.

This questionnaire was open from 2020-02-19 to 2020-02-25.

13 answers have been received.

Jump to results for question:

  1. Ok to publish the FPWD?

1. Ok to publish the FPWD?

The Silver Task Force and Silver Community Group believes that we have made substantial changes which address the prominent comments from the survey. However, we realize that some of the comments are complex and will need to be addressed more deeply in future Working Drafts. Some of the results in the Silver survey expressed opposing positions. Broader input would be useful to resolve them. Several of the complex issues will also benefit from broad public input.

The Silver Task Force and Silver Community Group would like to receive broad public input on a First Public Working Draft based on the Editor's Draft linked in this survey.

Do you agree to publish a First Public Working Draft based on the changes listed in this survey? If no, please briefly state what changes would have to be made for your approval.

Summary

ChoiceAll responders
Results
yes 6
no 7
concur
abstain

Details

Responder Ok to publish the FPWD?Comments
Luis Garcia yes
Bruce Bailey yes
Kim Dirks yes Yes, we need to get this out for wider review.
Rachael Bradley Montgomery yes The approach Silver is proposing is very different from the current approach to guidelines and it need to take a different approach in order to adequately meet the needs of individuals with cognitive and learning disabilities. I expect numerous questions and feedback from a public review but that is normal with change of this size, indicated by the movement from a 2.x to 3.0 release of the standards. I believe putting Silver's work out, with sample content, to begin the public conversation is valuable both because it will gather much needed input but also begin the conversation needed to lead up to this change. The content that is currently provided is substantive enough to gather meaningful comments and so I believe it should be published with publicity stressing the goals and limitations of the draft.
John Rochford yes
Jeanne F Spellman yes We have made changes where there is agreement. Many issues need broader public input to determine the correct direction. This is a right time to publish a FPWD.
Andrew Somers no Visual Contrast as stands (minus the pull request) shows incorrect math and some other issues. Unfortunately, due to a health issue coupled with problems with GitHub, I was unable to get the corrections out to Jeanne in time for the survey. The proposed pull request is viewable here : https://raw.githack.com/Myndex/silver/master/guidelines/methods/Method-font-characteristic-contrast.html

This shows substantial changes, and if this is "too much" of a change at this time, I do understand. HOWEVER, please be advised that there are errors in the math shown on "Detailed Description". If the proposed pull request is in too late for the first working draft, then I should at least make the corrections to the Detailed Description tab (I can do that immediately) perhaps as a separate interim pull request??

Sorry for any problems this is causing.
Wilco Fiers no See much of my previous feedback. In summary

- Normative parts must be testable statements, they aren't
- Removing accessibility support prioritises the needs of authors over the needs of users
- Scope goals seem to conflict each other and the scoring section. For example, better reflect the experience of users, and not taking impact or difficulty for users into account in scoring.
- There is insufficient content for a public review, most significantly the direction for the conformance model is insufficiently fleshed out.
- The document is not in the right template, and there is no indication of which section is normative and which is informative
- There are typos in the document

David MacDonald no > Created a detailed example of how a website could be scored (separate Google document)

I'm guessing its this document which I found linked in a Silver minutes email (suggest adding it here if its not linked in the Sliver Draft)
https://docs.google.com/document/d/1LfzTd_8WgTi0IUOOjUCRfRQ7e7__FRcnZow4w7zLlkY/edit

My understanding is that the scope of a claim is whatever the owner deems it is... so let's say its a sample of 30 pages from a 50,000 page site which the claim will apply to.
Are we going to document every instance of meaningful sequence, every instance of a heading, every instance of a form field, and link, and then subtract the number of errors against all of those found in the 30 pages?

Requiring a density score of total instances minus failures means we have to count all the passes of each Guideline AND all the failures. Documenting passes and counting those when automation is not possible seems like it will unnecessarily increase the evaluation time over the 2.x conformance model which just documents failures. For instance if there are 400 passes of meaningful sequences in 30 pages, who's going to count them or even determine what is a passed meaningful sequence... whereas an experienced 2.x evaluator will just skim the page and document places where the sequence is out. Seems like a huge additional burden on manual testers.
Alastair Campbell no I can see a great deal of improvement, but there are a few things left before it is ready for a wider review.

I've put all of them here for easier reading, a lot is editorial:
https://docs.google.com/document/d/1t0_AUOHVZ37P5VqYJnyZEJWGdO1CYw4x-THLe5l02K4/edit#

The 'substantive' ones are:

## Name

Due to charter comments/objections, I think we'll need to change the title from "Silver" to WCAG 3 throughout.

## Normative

> "Guidelines in plain language that are normative requirements and apply broadly across platforms. The specific tests to evaluate the Guidelines are in informative Methods."

I suspect this will be an issue, as the way that you pass or fail (however measured) is pretty much the key part of the guidelines. I'm happy to be corrected on this, but for example:
"Provide sufficient contrast between foreground text and its background."
Contrast is one of the more concrete requirements, but it is not a "normative requirements [you can] apply broadly across platforms."
You cannot apply 'sufficient' without more definition.

I think the way to deal with this (without hampering the progress) is to have a view of the guidelines that pulls in the test-methods (as 'includes' essentially), and have that be the 'normative' view of the spec.

## Exceptions

Headings includes an exception based on who is responsible for content ("You don't need to insert headings into documents you upload from another author").
I do not think this should be included unless we've had a reasonable discussion about how this would work.
Overall, I don't think we should define responsibilities at the method level, the exceptions should be based on the content/thing/work being done, not who did it.

Clear language has exceptions that are more about the content/work, but it is not very clear when something is an exception or not.

## Overlap between How-to and Methods

In the Clear language guideline, there is a lot of overlap between the guideline and the main method for that guideline "Edit text for clear language".
That is going to make it very hard to keep them exactly aligned, and as a user it feels like you are re-reading things that are sort of different but not really.

## Example needed

We need an example of working through the (3) current guidelines to produce a score, ideally in the explainer. Also, we need the explainer!
John Foliot no Wilco has captured Deque's concerns.
Laura Carlson no The statement "We propose changing the name of 'Color Contrast' to 'Visual Contrast'" is inaccurate. WCAG 2.X does not have a "Color Contrast" SC. It has SC 1.4.3 'Contrast (Minimum)' and SC 1.4.6 'Contrast (Enhanced)'. In WCAG 2.x, contrast is a measure of the difference in perceived "luminance" or brightness between two colors (the phrase "color contrast" is never used). Suggest restating to "We propose changing the names of 'Contrast (Minimum)' and 'Contrast (Enhanced)' to 'Visual Contrast of Text'.

I fear that "Allow organizations to decide the important parts of their products or sites to assess for accessibility" could be a way to game the system.
Andrew Kirkpatrick no I share many of the concerns raised around testability/conformance that need clarification.

In addition, the document is put forward as a Community Group Report and includes text like:
"This specification was published by the Silver Community Group. It is not a W3C Standard nor is it on the W3C Standards Track."
How will this be changed? I expect that this is being put forward by the AGWG as WCAG 3.0 FPWD.

More details on responses

  • Luis Garcia: last responded on 21, February 2020 at 19:23 (UTC)
  • Bruce Bailey: last responded on 21, February 2020 at 19:23 (UTC)
  • Kim Dirks: last responded on 21, February 2020 at 19:23 (UTC)
  • Rachael Bradley Montgomery: last responded on 21, February 2020 at 22:08 (UTC)
  • John Rochford: last responded on 22, February 2020 at 01:37 (UTC)
  • Jeanne F Spellman: last responded on 22, February 2020 at 18:25 (UTC)
  • Andrew Somers: last responded on 24, February 2020 at 04:09 (UTC)
  • Wilco Fiers: last responded on 24, February 2020 at 10:29 (UTC)
  • David MacDonald: last responded on 24, February 2020 at 22:21 (UTC)
  • Alastair Campbell: last responded on 25, February 2020 at 15:40 (UTC)
  • John Foliot: last responded on 25, February 2020 at 16:00 (UTC)
  • Laura Carlson: last responded on 25, February 2020 at 16:09 (UTC)
  • Andrew Kirkpatrick: last responded on 25, February 2020 at 16:23 (UTC)

Non-responders

The following persons have not answered the questionnaire:

  1. Gregg Vanderheiden
  2. Janina Sajka
  3. Shadi Abou-Zahra
  4. Makoto Ueki
  5. Peter Korn
  6. Léonie Watson
  7. David Sloan
  8. Mary Jo Mueller
  9. John Kirkwood
  10. Detlev Fischer
  11. Matt Garrish
  12. Chris Loiselle
  13. Sarah Horton
  14. JaEun Jemma Ku
  15. Francis Storr
  16. Aparna Pasi
  17. Ruoxi Ran
  18. Charles Adams
  19. Arthur Soroken
  20. David Fazio
  21. Daniel Montalvo
  22. Caryn Pagel
  23. Todd Libby
  24. Julia Chen
  25. Rain Breaw Michaels
  26. Jaunita George
  27. Suji Sreerama

Send an email to all the non-responders.


Compact view of the results / list of email addresses of the responders

WBS home / Questionnaires / WG questionnaires / Answer this questionnaire