W3C

List of comments on “Techniques for WCAG 2.0” (dated 16 January 2014)

Quick access to

There are 31 comments (sorted by their types, and the section they are about).

1-20 21-31

question comments

Comment LC-2874: Edits to G200
Commenter: Devarshi Pant <devarshipant@gmail.com> (archived message)
Context: in
assigned to Andrew Kirkpatrick
Resolution status:

G200: Opening new windows and tabs from a link only when necessary
url: http://www.w3.org/TR/WCAG-TECHS/G200.html

Under description:
A. What does 'outside of the secured scope' imply in #2?
B. 'It is recommend' should be 'It is recommended.'
(space separated ids)
(Please make sure the resolution is adapted for public consumption)

general comment comments

Comment LC-2894: Inaccessible supporting PDF examples
Commenter: Leona Zumbo <Leona.Zumbo@visionaustralia.org> on behalf of Vision Australia (archived message)
Context: in (PDF)
assigned to Andrew Kirkpatrick
Resolution status:

A number of the PDF example documents supporting the PDF techniques remain incorrect. Digital Access at Vision Australia have previously notified the W3C of the specific errors within these documents. The incorrect examples are likely to cause confusion for those that are new to implementing these techniques so we would suggest the examples are updated.
(space separated ids)
(Please make sure the resolution is adapted for public consumption)

Comment LC-2906: Technique F24 promotes non-existing "color" attribute of body in example #2 (#16)
Commenter: Webcc on behalf of Fraunhofer FIT
Context: in
assigned to Jonathan Avila
Resolution status:

In the file:

https://github.com/w3c/wcag/blob/master/wcag20/sources/failure-tech-src.xml

i.e., see:

http://www.w3.org/TR/WCAG-TECHS/F24.html

Example #2 is using a "color" attribute on "body" element, which does not exist in the specification. Please, modify it.
(space separated ids)
(Please make sure the resolution is adapted for public consumption)

Comment LC-2879: Move the logic away from How to meet and into the technique
Commenter: Wilco Fiers <w.fiers@accessibility.nl> on behalf of Accessibility Foundation (archived message)
Context: G101: Providing the definition of a word or phrase used in an unusual or r...
assigned to Loretta Guarino Reid
Resolution status:

G101 is not a solution to SC 3.1.3 on it’s own. The how to meet document contains a structure based on multiple possible combinations of techniques, including G101 to create a few distinct possible solutions to SC 3.1.3. This is fairly confusing and almost impossible to test. This is the only criterium in How to meet that works by pairing multiple techniques in different ways to gain a single result. Reading one of the techniques thus doesn’t really tell you how to solve for SC 3.1.3. I expect this was done to avoid duplication of content, but I think this little bit of duplication can avoid a lot of confusion. And it’s not like there is no duplication in other techniques either. So instead of building a logical structure in How to meet, move this logic into one or multiple techniques.

This comment is part of the project for the Accessibility Support Database

Proposed Change:
I think the best solution would be to flatten the How to meet into simply having the following 5 techniques, each of which can be used to meet the criteria without strange combinations with other techniques.
- H40: Using definition lists
- H60: Using the link element to link to a glossary
- H54: Using the dfn element to identify the defining instance of a word
- G62: Providing a glossary
- G70: Providing a function to search an online dictionary

All the things the other criteria require is moved into these techniques, such as that with a definition list you should also link the definition to the definition list, and that if the same phrase is used differently on the same page it is insufficient to only link the first occurrence. This could perhaps also be a failure.
(space separated ids)
(Please make sure the resolution is adapted for public consumption)

Comment LC-2881: This technique should be advisory G183
Commenter: Wilco Fiers <w.fiers@accessibility.nl> on behalf of Accessibility Foundation (archived message)
Context: G183: Using a contrast ratio of 3:1 with surrounding text and providing ad...
assigned to Andrew Kirkpatrick
Resolution status:

This technique is a great idea, but it’s not sufficient to meet success criterion 1.4.1. There is no allowance in 1.4.1 for information to be conveyed through color alone as long as those colors have a large contrast ratio. There is simply nothing in the criterion that mentions any such thing. So even though this makes sense as a technique, this exception is not permitted by the normative part of WCAG and so it can not be considered sufficient. Since this is still a pretty good idea, it’s best to make this technique an advisory technique.

Proposed Change:
(space separated ids)
(Please make sure the resolution is adapted for public consumption)

Comment LC-2880: This technique should be a failure technique
Commenter: Wilco Fiers <w.fiers@accessibility.nl> on behalf of Accessibility Foundation (archived message)
Context: H2: Combining adjacent image and text links for the same resource
assigned to Loretta Guarino Reid
Resolution status:

As an HTML technique, this one is very strange. It doesn't give you a way to solve a certain problem, rather it says not to do something; namely to have two adjacent links with the same description. This seems much more the kind of thing failure techniques are for, the "Don't do X"-type. Failing H2 doesn't mean you failed the SC, since you can still have two links, one with an image and the next with a text, and the image has the same alternative as the text. The image can still meet technique H37 (img with descriptive alt) and thus someone might conclude this meets the success criteria.

There is a pretty good argument that can be made against this scenario. If the W3C logo has the text "W3C logo" adjacent to it, this could be considered it's text alternative. Giving it an alt text of "W3C logo" would be redundant and thus the combined result of the two alternatives would be "W3C logo W3C logo" which is quite clearly not a good alternative. When these two bits of content are in separate links however, the content author MUST provide a text for each link in order to meet 4.1.2. So leaving the alt attribute empty wouldn't be a solution in this case either. The only possible solution would be to combine the two links.

If this argument is valid (And I believe WCAG isn't quite specific enough to decide either way, but opinions my vary on this.), then that would mean having adjacent links one with an image and the other with a text would that had the same description would always be a failure. For 1.1.1 if the image repeated the text in its alt attribute and for 4.1.2 if the alt attribute was left empty.

This comment is part of the project for the Accessibility Support Database
(space separated ids)
(Please make sure the resolution is adapted for public consumption)

Comment LC-2890: This technique entirely covered by H53
Commenter: Wilco Fiers <w.fiers@accessibility.nl> on behalf of Accessibility Foundation (archived message)
Context: H53: Using the body of the object element
assigned to Kathleen Anderson
Resolution status:

This technique seems to be similar to technique H53, both use the body of an object element. Technique H27 seems incomplete however. It doesn’t have any tests and it’s examples doesn’t cover anything that H53 doesn’t cover. Perhaps H27 or H53 should be removed or the two should be merged.

This comment is part of the project for the Accessibility Support Database.
(space separated ids)
(Please make sure the resolution is adapted for public consumption)

Comment LC-2882: Add a warning about the poor support for this technique h69
Commenter: Wilco Fiers <w.fiers@accessibility.nl> on behalf of Accessibility Foundation (archived message)
Context: H69: Providing heading elements at the beginning of each section of content
assigned to Sailesh Panchang
Resolution status:

It seems pretty important to note that most user agents do not support navigating between headings, and not a single one support navigating frames (H70) or maps (H50). To my understanding that means that in practice none of these techniques can be relied upon except in the extremely rare situation where someone is on a closed network using Opera (which has no screen reader support as far as I’m aware). It seems to me that a warning for this would be in order.

This comment is part of the project for the Accessibility Support Database.
(space separated ids)
(Please make sure the resolution is adapted for public consumption)

Comment LC-2889: The technique should be more specific h86
Commenter: Wilco Fiers <w.fiers@accessibility.nl> on behalf of Accessibility Foundation (archived message)
Context: H86: Providing text alternatives for ASCII art, emoticons, and leetspeak
assigned to Joshue O Connor
Resolution status:

This technique is quite vague and seems to cover multiple solutions. The test only requires test before or after the ASCII graphic, but in the examples you see much more then that. Including a solution with the abbr element that doesn’t actually pass the test. This technique might need to be rethought. Perhaps it can be a general technique?

This comment is part of the project for the Accessibility Support Database.
(space separated ids)
(Please make sure the resolution is adapted for public consumption)

Comment LC-2887: This technique should be generalized h87
Commenter: Wilco Fiers <w.fiers@accessibility.nl> on behalf of Accessibility Foundation (archived message)
Context: H87: Not interfering with the user agent's reflow of text as the viewing w...
assigned to Andrew Kirkpatrick
Resolution status:

This technique doesn't test for anything specific in HTML. Many technologies can reflow the way is described in this technique. Making this a general technique would mean it would be applicable to these technologies.

This comment is part of the project for the Accessibility Support Database.
(space separated ids)
(Please make sure the resolution is adapted for public consumption)

Comment LC-2884: This technique should be generalized h90
Commenter: Wilco Fiers <w.fiers@accessibility.nl> on behalf of Accessibility Foundation (archived message)
Context: H90: Indicating required form controls using label or legend
assigned to Sailesh Panchang
Resolution status:

This technique doesn't test for anything specific in HTML. What the technique does is ensure that the indication that a field is required is part of a label element. This can very easily be generalized by stating that this indication should be part of the label (which may or may not be an html label element). That would make the technique more flexible and usable in other technologies.

This comment is part of the project for the Accessibility Support Database.
(space separated ids)
(Please make sure the resolution is adapted for public consumption)

Comment LC-2877: The technique should mention input type = file
Commenter: Wilco Fiers <w.fiers@accessibility.nl> (archived message)
Context: H91: Using HTML form controls and links
assigned to Andrew Kirkpatrick
Resolution status:

The description contains a table with the different input elements. The table should also include <input type = "file"> in one of the rows.

This comment is part of the project for the Accessibility Support Database
(space separated ids)
(Please make sure the resolution is adapted for public consumption)

Comment LC-2885: This technique should be generalized h92
Commenter: Wilco Fiers <w.fiers@accessibility.nl> on behalf of Accessibility Foundation (archived message)
Context: H92: Including a text cue for colored form control labels
assigned to Marc Johlic
Resolution status:

This technique doesn’t test for anything specific in HTML. The proposed solution in this technique can be applied in any technology. There is actually nothing in this technique other then it’s identifier that would have to change for this technique to be a general technique. Making this a general technique would mean it would be applicable to other technologies.

This comment is part of the project for the Accessibility Support Database.
(space separated ids)
(Please make sure the resolution is adapted for public consumption)

Comment LC-2883: HTML technique is a duplicate of h93
Commenter: Tom Siechert <tsiechert@csufresno.edu> on behalf of California State University Fresno (archived message)
Context: H93: Ensuring that id attributes are unique on a Web page
assigned to Andrew Kirkpatrick
Resolution status:

It appears that this HTML technique is a duplicate of http://www.w3.org/TR/2005/WD-WCAG20-HTML-TECHS-20051123/#H93, however, this H93 is of a different subject matter compared to the H93 referenced in the URL above.

Proposed Change:
Change in numbering of this (or the other H93) to keep both techniques as their own separate technique.
(space separated ids)
(Please make sure the resolution is adapted for public consumption)

Comment LC-2888: This technique should be generalized C26
Commenter: Wilco Fiers <w.fiers@accessibility.nl> on behalf of Accessibility Foundation (archived message)
Context: C26: Providing options within the content to switch to a layout that does ...
assigned to Marc Johlic
Resolution status:

This technique doesn’t test for anything specific to CSS. Providing a way to change the layout of a web page is not something that can exclusively be done in CSS either. Most technologies can do it. Making this a general technique would mean it would be applicable to other technologies.

This comment is part of the project for the Accessibility Support Database.
(space separated ids)
(Please make sure the resolution is adapted for public consumption)

Comment LC-2886: This technique should be generalized C29
Commenter: Wilco Fiers <w.fiers@accessibility.nl> on behalf of Accessibility Foundation (archived message)
Context: C29: Using a style switcher to provide a conforming alternate version
assigned to Marc Johlic
Resolution status:

This technique doesn’t test for anything specific to CSS. Changing the style of components on a web page using some kind of style switch can be done in any technology using the method described in this technique. Making this a general technique would mean it would be applicable to other technologies.

This comment is part of the project for the Accessibility Support Database.
(space separated ids)
(Please make sure the resolution is adapted for public consumption)

Comment LC-2891: The technique doesn't solve the problem of the success criterion SCR21
Commenter: Wilco Fiers <w.fiers@accessibility.nl> on behalf of Accessibility Foundation (archived message)
Context: SCR21: Using functions of the Document Object Model (DOM) to add content t...
assigned to Joshue O Connor
Resolution status:

This technique is a clear “Don’t do X”, in this case document.write and innerHTML. Which would make it a failure technique, except that using document.write / innerHTML is more a question of accessibility support then of conformance to any specific criterion. I imagine this technique was created to inform developers about the innerHTML problem. But it’s done in a format that seems inappropriate for it; as a script technique.

For all techniques content can be generated using innerHTML, so limiting this to 1.3.1 seems odd. Also, this technique doesn’t actually provide a solution to 1.3.1. If you were to insert a heading without header markup, that would be an accessibility problem, even if it was done using DOM. It would probably be more appropriate to scratch this technique as it is and create an article in which this problem is explained. This article can then be referenced to from other documents.

This comment is part of the project for the Accessibility Support Database.
(space separated ids)
(Please make sure the resolution is adapted for public consumption)

Comment LC-2903: Technique F47 improvement
Commenter: Webcc on behalf of Fraunhofer FIT
Context: F47: Failure of Success Criterion 2.2.2 due to using the blink element
assigned to Jonathan Avila
Resolution status:

In the file:

https://github.com/w3c/wcag/blob/master/wcag20/sources/failure-tech-src.xml

The technique F47 should make a reference to the own documentation of Mozilla, where the use of blink is deprecated and discouraged:

https://developer.mozilla.org/en-US/docs/Web/HTML/Element/blink
(space separated ids)
(Please make sure the resolution is adapted for public consumption)

Comment LC-2892: F76: Failure of Success Criterion 3.2.2
Commenter: Devarshi Pant <devarshipant@gmail.com> (archived message)
Context: F76: Failure of Success Criterion 3.2.2 due to providing instruction mater...
assigned to Andrew Kirkpatrick
Resolution status:

Refer to the failure technique, 'F76: Failure of Success Criterion 3.2.2
due to providing instruction material about the change of context by change
of setting in a user interface element at a location that users may bypass'
at http://www.w3.org/TR/WCAG-TECHS/F76.html.

Note the title of the failure technique and then note that the failure is
not due to providing instructional material but due to lack thereof.

Suggested change: 'F76: Failure of Success Criterion 3.2.2 due to lack of
instruction material about the change of context by change of setting in a
user interface element at a location that users may bypass.'

-Devarshi
(space separated ids)
(Please make sure the resolution is adapted for public consumption)

typo comments

Comment LC-2875: Typo in G192
Commenter: Devarshi Pant <devarshipant@gmail.com> (archived message)
Context: in
assigned to Joshue O Connor
Resolution status:

G192: Fully conforming to specifications

URL: http://www.w3.org/TR/WCAG-TECHS/G192.html

Under Examples, correct the sentence "It is run though ..." to "It is run
through ...."
(space separated ids)
(Please make sure the resolution is adapted for public consumption)

1-20 21-31

Add a comment.


Developed and maintained by Dominique Hazaël-Massieux (dom@w3.org).
$Id: Overview.php,v 1.47 2015-01-05 13:17:17 dom Exp $
Please send bug reports and request for enhancements to w3t-sys.org