New rule: Image contains no text (0va7u6) #1522

carlosapaduarte · 2020-12-16T16:42:49Z

New rule for 1.4.5 checking that images do not contain text unless it is essential

Need for Final Call:
This will require a 2 weeks Final Call (new rule)

Pull Request Etiquette

When creating PR:

Make sure you're requesting to pull a branch (right side) to the develop branch (left side).
Make sure you do not remove the "How to Review and Approve" section in your pull request description

After creating PR:

Add yourself (and co-authors) as "Assignees" for PR.
Add label to indicate if it's a Rule, Definition or Chore.
Link the PR to any issue it solves. This will be done automatically by referencing the issue at the top of this comment in the indicated place.
Optionally request feedback from anyone in particular by assigning them as "Reviewers".

When merging a PR:

Close any issue that the PR resolves. This will happen automatically upon merging if the PR was correctly linked to the issue, e.g. by referencing the issue at the top of this comment.

How to Review And Approve

Go to the “Files changed” tab
Here you will have the option to leave comments on different lines.
Once the review is completed, find the “Review changes” button in the top right, select “Approve” (if you are really confident in the rule) or "Request changes" and click “Submit review”.
Make sure to also review the proposed Final Call period. In case of disagreement, the longer period wins.

…ub.io into develop

_rules/image-no-text-0va7u6.md

Jym77 · 2020-12-22T08:05:26Z

_rules/image-no-text-0va7u6.md

+
+## Assumptions
+
+The text is the most significant content in the image. If there is text in the image, but it is not the most significant content, this rule might fail but the success criterion might still be satisfied.


Not sure why this is an assumption rather than part of the Expectation. Do you feel it's ambiguous? (it certainly is subjective).
I guess this is to exclude, say, image of a street with street signs or shop signs showing text. I'm a bit afraid this is going to exclude a ton of images from the rule (i.e. that it is a very strong assumption that often won't hold).

Yes, I do feel "most significant content" is ambiguous. That's why I moved this to an assumption.

It would be nice to have input on this from other reviewers...

The WCAG definition for "image of text" already excludes "text that is part of a picture that contains significant other visual content", so I'm not sure we need to have this?

But we don't link to the WCAG definition of "image of text" so we need to have this here (or worked into the expectation).
And I don't think we could use it, because, at least for me, the part about "significant other visual content" is ambiguous. If you consider the example given in the definition, an image of "a person's name on a name tag in a photograph", assuming that the person is also on the photograph, I wouldn't consider it to be an image of text because there would be other significant visual content on the image.. So, I guess, this is ambiguous...

I don't quite see what is ambiguous about "significant other visual content". Can you explain what different ways there are to interpret this phrase? Sure we can disagree over when something is "significant" in the same way we can disagree about if something is "descriptive". There's a quantitative aspect to it; How significant is enough? How descriptive is enough? But I think the meaning of this is clear. It's subjective, but I don't see this is ambiguous.

My main concern would not be the "how much significant" but the "is this part of the image significant". As I said in my previous reply, the example of "image of text" in the WCAG definition is not something that I would consider an image of text (it's true that the example needs context that is not provided, so I had to assume the context), so I have to conclude that the "significant other visual content" is being ambiguously interpreted.

Nevertheless, I seem to be the only one having this doubt, so I can give it a try and include this in the expectation.

_rules/image-no-text-0va7u6.md

updates plus discussion needed

Jym77

Couple new questions…

_rules/image-no-text-0va7u6.md

Jym77 · 2020-12-23T12:50:05Z

_rules/image-no-text-0va7u6.md

+
+## Applicability
+
+This rule applies to any HTML element that is [visible][] for which one of the following is true:


Should there be a mention somewhere that the image is not decorative? ("convey information")

Either by checking roles rather than element (can be in applicability), or adding a condition in expectation about "the image is [pure decoration][]".

This was one of the the first thoughts that crossed my mind when I started thinking about this rule. I discussed this with @WilcoFiers and he easily convinced me that this is not relevant for this SC. If you are presenting text then you should use text and not images of text unless the presentation is essential. You can always mark text as decorative if it is not being used to convey information.

🤔 I have to disagree with the conclusion…
1.4.9, the corresponding AAA criterion, is explicitly allowing images of text that are pure decoration. It would be weird if they were failing 1.4.5…

That line in 1.4.9 really made me change my mind. I've updated the applicability.

_rules/image-no-text-0va7u6.md

Co-authored-by: Jean-Yves Moyen

now with background images

Jym77 · 2021-01-04T12:12:12Z

_rules/image-no-text-0va7u6.md

+
+## Applicability
+
+This rule applies to any HTML element that is [visible][] for which one of the following is true:


🤔 I have to disagree with the conclusion…
1.4.9, the corresponding AAA criterion, is explicitly allowing images of text that are pure decoration. It would be weird if they were failing 1.4.5…

Jym77 · 2021-01-04T12:14:13Z

_rules/image-no-text-0va7u6.md

@@ -26,11 +26,12 @@ acknowledgments:
 This rule applies to any HTML element that is [visible][] for which one of the following is true:

 - the element is an `img` element where at least one of the [image sources][] in its [source set][] does not reference an SVG document; or
- the element is an `input` element in the [Image Button][] state and its `src` [attribute value][] does not reference an SVG document.
+- the element is an `input` element with a `type` [attribute value][] of `image` and its `src` [attribute value][] does not reference an SVG document; or
+- the element has a [`background-image`][background-image] CSS property with at least one [`image`][css-image] that is a [url reference][url-reference] that does not reference an SVG document.


Which value of the property do you consider? I guess the computed value (after resolving cascade and such).

_rules/image-no-text-0va7u6.md

…t-rules/act-rules.github.io into image-contains-no-text-0va7u6

carlosapaduarte · 2021-01-20T13:11:06Z

@nitedog ready for your review again (changes made to the examples)

WilcoFiers · 2021-01-22T11:33:14Z

_rules/image-no-text-0va7u6.md

- the image resource contains text and it is [essential][] that the text is rendered with that specific presentation; or
- the image resource does not contain text expressing anything in a [human language][]; or
- the image resource contains text and the text is not a significant part of the visual content in the image; or
+- the [visible pixels][visible] of the rendered image resource contain text and it is [essential][] that the text is rendered with that specific presentation; or


This seems like it could be simplified:

Suggested change

- the [visible pixels][visible] of the rendered image resource contain text and it is [essential][] that the text is rendered with that specific presentation; or

- the [visible pixels][visible] of the rendered image resource contain text for which its presentation is [essential]; or

WilcoFiers · 2021-01-22T11:37:51Z

_rules/image-no-text-0va7u6.md

@@ -36,24 +36,23 @@ This rule applies to any [visible][] [embedded image][].

 For the rendered image resource of the [image sources][] of each test target, at least one of the following is true:


I think we should hoist the visible pixels thing into the setup to avoid repeating that phrase 3 times:

Suggested change

For the rendered image resource of the [image sources][] of each test target, at least one of the following is true:

For the [visible pixels][] of the rendered image resource of the [image sources][] of each test target, at least one of the following is true:

The issue is that there is a 4th condition...

pages/glossary/embedded-image.md

…d image resource

changes made

Jym77

I think the wordcloud asset is not used anymore and can be deleted.

Jym77 · 2021-01-26T07:55:28Z

_rules/image-no-text-0va7u6.md

+For each test target, at least one of the following is true:

- the [visible pixels][visible] of the rendered image resource contain text and it is [essential][] that the text is rendered with that specific presentation; or
- the [visible pixels][visible] of the rendered image resource do not contain text expressing anything in a [human language][]; or
- the [visible pixels][visible] of the rendered image resource contain text and the text is not a significant part of the visual content in the image; or
- the test target is [purely decorative][].
+- the [visible pixels][visible] of the test target contain text for which its presentation is [essential][]; or
+- the [visible pixels][visible] of the test target do not contain text expressing anything in a [human language][]; or
+- the [visible pixels][visible] of the test target contain text and the text is not a significant part of the [visible pixels][visible] of the test target; or
+- the [embedded image][] the test target belongs to is [purely decorative][].


I guess if we want to factor in the "visible pixels" bit, we could do something like

For each test target, either is belongs to a purely decorative image or at least on of the following is true for its visible pixels:

But I'm not sure this is much better because it extracts a condition from the list, and because we usually want each individual condition to actually frontload the common thingie they apply to, so they would still real "the visible pixels contains no text", …

I would prefer to keep it as a list with all the conditions for the reasons you enumerate. If others prefer this solution, I won't oppose to the change.

_rules/image-no-text-0va7u6.md

Co-authored-by: Jean-Yves Moyen

nitedog

There are still some comments on the definition for Embedded Image but I understand that these are mostly editorial.

carlosapaduarte · 2021-01-27T12:31:36Z

Final call until February 10

carlosapaduarte · 2021-02-15T13:12:46Z

Final call ended. Merging

carlosapaduarte added 3 commits December 16, 2020 16:39

Create image-no-text.md

390541a

add resources

3abfbf6

Merge branch 'develop' of https://github.com/act-rules/act-rules.gith…

8cfea61

…ub.io into develop

carlosapaduarte changed the title ~~Image contains no text 0va7u6~~ New rule: Image contains no text (0va7u6) Dec 16, 2020

carlosapaduarte self-assigned this Dec 16, 2020

carlosapaduarte added reviewers wanted Rule Use this label for a new rule that does not exist already labels Dec 16, 2020

carlosapaduarte added 2 commits December 16, 2020 17:51

update file name to include id

3342c71

Merge branch 'develop' into image-contains-no-text-0va7u6

3256066

Jym77 previously requested changes Dec 22, 2020

View reviewed changes

carlosapaduarte added 3 commits December 23, 2020 10:09

Merge branch 'develop' into image-contains-no-text-0va7u6

b1df96a

add background note explaining applicability

9264311

include multiple image sources

bf572d1

carlosapaduarte requested a review from Jym77 December 23, 2020 11:12

Jym77 previously requested changes Dec 23, 2020

View reviewed changes

carlosapaduarte and others added 5 commits December 27, 2020 09:26

Update _rules/image-no-text-0va7u6.md

5d45fbf

Co-authored-by: Jean-Yves Moyen

Update _rules/image-no-text-0va7u6.md

25a7598

Co-authored-by: Jean-Yves Moyen

include background-images in the applicability

2652791

update inapplicable example 3 to consider all applicable elements

cc202a0

update applicability of background-images

6dd54a4

carlosapaduarte requested review from Jym77 and WilcoFiers December 27, 2020 10:01

Jym77 previously requested changes Jan 4, 2021

View reviewed changes

carlosapaduarte added 5 commits January 4, 2021 16:24

Merge branch 'develop' into image-contains-no-text-0va7u6

e9da6d6

make decorative images not applicable

2436036

Merge branch 'image-contains-no-text-0va7u6' of https://github.com/ac…

04d7869

…t-rules/act-rules.github.io into image-contains-no-text-0va7u6

add computed style plus update links

2782ad0

add sc1.4.9

688c7e0

carlosapaduarte added 3 commits January 20, 2021 12:49

fix tests

ebe179d

remove unused resources

65168c8

fix tests

5541bfa

carlosapaduarte requested a review from nitedog January 20, 2021 13:10

nitedog previously approved these changes Jan 20, 2021

View reviewed changes

WilcoFiers previously requested changes Jan 22, 2021

View reviewed changes

carlosapaduarte added 3 commits January 25, 2021 16:03

Create rendered-image-resource.md

91c050c

update applicability and expectation to use the definition of rendere…

7165b03

…d image resource

update example descriptions

3234ce4

carlosapaduarte requested review from WilcoFiers, nitedog and Jym77 January 25, 2021 16:17

fix test

2128796

Jym77 approved these changes Jan 26, 2021

View reviewed changes

carlosapaduarte and others added 2 commits January 26, 2021 09:26

Update _rules/image-no-text-0va7u6.md

e476823

Co-authored-by: Jean-Yves Moyen

Delete wordcloud.png

c13750f

WilcoFiers approved these changes Jan 26, 2021

View reviewed changes

nitedog approved these changes Jan 27, 2021

View reviewed changes

Merge branch 'develop' into image-contains-no-text-0va7u6

7c2be12

carlosapaduarte added Review call 2 weeks Call for review for new rules and big changes and removed reviewers wanted labels Jan 27, 2021

Merge branch 'develop' into image-contains-no-text-0va7u6

c752dd1

carlosapaduarte merged commit f1b55d3 into develop Feb 15, 2021

carlosapaduarte deleted the image-contains-no-text-0va7u6 branch February 15, 2021 13:12


		## Assumptions

		The text is the most significant content in the image. If there is text in the image, but it is not the most significant content, this rule might fail but the success criterion might still be satisfied.


		## Applicability

		This rule applies to any HTML element that is [visible][] for which one of the following is true:

	- the [visible pixels][visible] of the rendered image resource contain text and it is [essential][] that the text is rendered with that specific presentation; or
	- the [visible pixels][visible] of the rendered image resource contain text for which its presentation is [essential]; or

		@@ -36,24 +36,23 @@ This rule applies to any [visible][] [embedded image][].

		For the rendered image resource of the [image sources][] of each test target, at least one of the following is true:

	For the rendered image resource of the [image sources][] of each test target, at least one of the following is true:
	For the [visible pixels][] of the rendered image resource of the [image sources][] of each test target, at least one of the following is true:

New rule: Image contains no text (0va7u6) #1522

New rule: Image contains no text (0va7u6) #1522

Uh oh!

Conversation

carlosapaduarte commented Dec 16, 2020 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Pull Request Etiquette

When creating PR:

After creating PR:

When merging a PR:

How to Review And Approve

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Jym77 left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

carlosapaduarte commented Jan 20, 2021

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Jym77 left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

nitedog left a comment

Choose a reason for hiding this comment

Uh oh!

carlosapaduarte commented Jan 27, 2021

Uh oh!

carlosapaduarte commented Feb 15, 2021

Uh oh!

Uh oh!

carlosapaduarte commented Dec 16, 2020 •

edited

Loading