SC 1.4.1 Use of Color and link identification in a data table

[sarasuri]

Hello,
I have some questions regarding the use of colour to identify links within a data table, specifically related to SC 1.4.1.
In a data table, for example the first column contains links (#1d659c), while the other columns display regular text (#000000). These links are visually differentiated from the surrounding text by colour alone, without underline or other visual cues.

Scenario 1:
If colour alone is used as the sole visual indicator to show that the “Product number” column contains links, and there are no other distinguishing indicators by default and the color contrast between the link and the surrounding text is less than 3:1, would this be considered a failure of SC 1.4.1?

Scenario 2:
If the colour contrast ratio between the link (#3366CC) and the surrounding text (#000000) meets the minimum 3:1 requirement, but no other visual cues (such as underline or bold formatting) are provided by default, would this pass the SC 1.4.1?

Scenario 3:
Does the WCAG requirement for using colour to indicate links within a block of text also apply to links in a data table, considering a table is a structured layout rather than continuous text?

Any insights or feedback on these questions would be greatly appreciated.
Thank you,

[patrickhlauke]

Scenario 1: fails 1.4.1
Scenario 2: passes 1.4.1
Scenario 3: there is no "WCAG requirement for using colour to indicate links within a block of text".

[sarasuri]

@patrickhlauke, thank you for your feedback.

For scenario 3, I meant the following:
"Does the WCAG requirement that color not be used as the only visual means of indicating links within a block of text also apply to links in a data table, considering that a table is a structured layout rather than continuous text?"

However, based on your feedback on Scenarios 1 and 2, I understand that the requirement regarding colour being the only visual means to indicate links within blocks of text also applies to data tables. In other words, even though a table is a structured format, the same rule applies: colour must not be the only visual indicator used to identify links.

The WCAG definition of “blocks of text” (https://www.w3.org/TR/WCAG22/#dfn-blocks-of-text) is what led to questions about whether content in tables counts as blocks of text.

[patrickhlauke]

The WCAG definition of “blocks of text” (https://www.w3.org/TR/WCAG22/#dfn-blocks-of-text) is what led to questions about whether content in tables counts as blocks of text.

that definition is irrelevant though because the normative wording of the success criterion (which is what matters) is not scoped/limited to "blocks of text" https://www.w3.org/TR/WCAG22/#use-of-color ... and even the non-normative understanding document doesn't mention blocks of text https://www.w3.org/WAI/WCAG22/Understanding/use-of-color.html

so again, there is no "WCAG requirement that color not be used as the only visual means of indicating links within a block of text"

[GreggVan]

Hmmm interesting question. I think it would be BEST if the header said (incl link) as a cue.

since pointing to them will also show them to be links -- it isnt color alone...... but it is visually...

If they were much lighter or bold then it wouldnt be color alone (it would also be lightness or boldness). But since titles are often bold that is not much of a cue.

I would say it was borderline pass (since it is lighter - but not by much) but very bad practice.

[patrickhlauke]

I would say it was borderline pass (since it is lighter - but not by much)

we did clarify a while ago that for use of color, we are guided by the 3:1 ratio (see second note in intent https://www.w3.org/WAI/WCAG22/Understanding/use-of-color.html) so I'd say not just best practice