W3C

Site Navigation

Nearby

Results of Questionnaire Error Prevention Guideline

The results of this questionnaire are available to anybody.

This questionnaire was open from 2021-08-12 to 2021-08-17.

12 answers have been received.

Jump to results for question:

Approval to move to CFC

1. Approval to move to CFC

Can we move the revised Error Prevention Guideline to CFC for addition to the next update of WCAG 3?

Summary

Choice	All responders
Choice	Results
Agree	8
Agree with the following changes	3
Disagree for the following reason	1

Details

Responder	Approval to move to CFC	Comments
Charles Adams	Agree
Laura Carlson	Agree
Ben Tillyer	Agree
Jeanne F Spellman	Agree
Sarah Horton	Agree
Jake Abma	Agree with the following changes	As a first drafty draft what we're working on is fine, the maturity is not completely as expected yet though.
Oliver Keim	Agree
Marc Johlic	Agree
Michael Gower	Disagree for the following reason	I understand that we have to balance a need for content with a need for a quality threshold, but I'm not sure that balance is being achieved: I find the Method "Instructions available at the source of input" needs a fair bit of work. Just a few examples of considerations from the introduction tab: >Data entry instructions that are persistent remain visible as users enter data. If they are persistent, by definition, they remain visible. Remove the words "that are persistent"? I also think we need to differentiate between important information and trivial placeholder text. A label named First Name with a placeholder "Enter your first name" should not be in violation of persistence; the placeholder text is superfluous. I suggest "Necessary data entry instructions remain visible as users enter data" >Instructions that display at the source of input are visible as users enter data. "At the source" is cryptic. And this phrase does not suggest to me a proximity requirement, which I think this is referring to? Suggest something like "Instructions are located in close proximity to user data entry points" > Instructions that are programmatically defined and associated with inputs can be conveyed by assistive technology as users enter data. The phrase "as users enter data" is fairly problematic. I think you mean they are available WHERE users enter data? This phrase also seems like it's both stating the problem and solving it, which seems a common way these are all phrased, reviewing them, but if intended that way, maybe these should be called 'Desired outcomes' or something? But then the How it solves user needs section would appear to be redundant? I feel like the template needs a once over. > Instructions that are relevant to associated inputs enable users to understand how to enter data correctly. This seems to be very redundant with another method, Instructions for completing tasks. One of the biggest challenges with 2.x IMO is that the requirements are not clearly differentiated (normalized, in the data modelling sense). Thus a tester can be unclear on which requirement is being failed. I'd really like to eliminate this kind of overlap wherever possible in 3.x -- Moving on to the Examples page, the second example 'Providing instructions outside labels' confuses technology and location. There should be no mention of aria-describedby, both because it is a solution for another method (Programmatically defined and associated instructions) and because it seems to morph "at the source" (or maybe I'm not understanding what you mean by this). I would have thought this would cover things like the location of instructions (possible placements). -- On the test tab, going to have to be super careful with the phrase "adjacent to the input". What if the instruction precedes the label?
Wilco Fiers	Agree with the following changes	1. The critical error, which is normative text refers to a test "Input instructions provided", which is an informative text. Normative, which have to be unchanging, can not be built on text that can be changed at any point in time. 2. There is no indication in normative text on how to come to a percentage score. As above, normative text can not rely on informative text. I can accept this if we add an editors note acknowledging that we still need to work out how to put the entire requirement in normative text.
Bruce Bailey	Agree
John Foliot	Agree with the following changes	+1 to Wilco's point.