RE: Unicode characters used as images

My opinion: you can argue the same thing for just about any text used to give a certain look. “It’s just text”. If you have a button with “x” that means “close this dialog”, the look and placement of the button tell sighted users what it is. But it’s not an equivalent experience for all users. I think that’s why it requires some other way to make it usable by everybody.

It’s probably worth looking closer at what WCAG says is “non-text content”: “any content that is not a sequence of characters that can be programmatically determined or where the sequence is not expressing something in human language. Note: This includes ASCII Art<https://www.w3.org/WAI/WCAG21/Understanding/non-text-content.html#dfn-ascii-art> (which is a pattern of characters), emoticons, leetspeak (which uses character substitution), and images representing text.”

If leetspeak can fall under this (e.g., spelling “cow” “c0w”), I think you can call a unicode glyph non-text content.

Personally I’d defer to Patrick on something semi-vague like this. So much of WCAG is subjective that you have to defer to people who’ve been around for a while sometimes, even when it’s not a cut-and-dry situation.

From: Pyatt, Elizabeth J <ejp10@psu.edu>
Sent: Tuesday, February 6, 2024 06:51
To: Léonie Watson <lwatson@tetralogical.com>
Cc: S <Starry_sky@live.com>; Ms J <ms.jflz.woop@gmail.com>; w3c-wai-ig <w3c-wai-ig@w3.org>
Subject: Re: Unicode characters used as images

Having monitored this thread, I would observe that some characters just have multiple uses. These include x (and also times/close/delete), the . (period/decimal point), * (star/times), and the - (hyphen/minus).

Sighted users are able to unconsciously distinguish the meaning based on the surrounding text. I recognize that this could be more difficult on a screen reader for a number of reasons.  Encoding the same shape to have different meanings is one way to account for this, but it probably won't account for every case such as the >  (greater than/forward arrow variant).


I'm not sure what the solution is, but I do miss having an ARIA Alt mechanism sometimes. Maybe some sort of shape key?


Thinking out loud.


Elizabeth



On Feb 6, 2024, at 2:38 AM, Léonie Watson <lwatson@tetralogical.com<mailto:lwatson@tetralogical.com>> wrote:

You don't often get email from lwatson@tetralogical.com<mailto:lwatson@tetralogical.com>. Learn why this is important<https://urldefense.com/v3/__https:/aka.ms/LearnAboutSenderIdentification__;!!C5qS4YX3!Exyu9WjqYP1mL2R2D8DYnkgwh2Ml3vCtw9Es-00dDFc6g20Kov9NqqirkhyLd85G6PX8xnCbUYgwRcpB$>
It might be argued that "heavy multiplication X" is not the most user-friendly way to indicate that an item has been checked off a list, but I don't believe it fails any WCAG SC. The way Unicode characters are announced by screen readers is not always as useful as we might like, but they are announced and the names they're given are usually understandable if less than perfectly so.

On 05/02/2024 21:44, S wrote:
But that "heavy x" is not an image or ascii art and will not be interpreted that way regardless of the intent. And, it would be confusing to mark up text as if it was an image.  Suggestion is to use text "Y" and "N" as indicators for better cognitive recognition.  But if it has to appear as "x" for visual effect, then they should use an actual image with valid alt text so it is recognized accordingly.
On 2/5/2024 7:53 AM, Ms J wrote:
Hello

If I had a shopping list and each item had a 'cross' next to it to indicate it was completed but the unicode 'heavy multiplication x' character was used for the cross, is this a failure of any sc?

They're basically using a text character like an image because they're using it for its physical characteristics, but they're not marking it up as an image (for example with an aria img role) and giving it an alt. It's read with JAWS as 'heavy multiplication x'.

My thoughts are - could it fail name, role, value because it's used as an image but doesnt have that role?
Could it fail info and relationships because it coveys information visually but not programmatically? (But then thats like saying images with unclear alts should fail 1.3.1)
I don't think it fails sensory characteristics because there's no corresponding instructions that refer to it by its appearance

Thanks

Sarah

Sent from Outlook for iOS<https://urldefense.com/v3/__https:/aka.ms/o0ukef__;!!C5qS4YX3!Exyu9WjqYP1mL2R2D8DYnkgwh2Ml3vCtw9Es-00dDFc6g20Kov9NqqirkhyLd85G6PX8xnCbUQF0iTlr$>




--

Léonie Watson (she/her)

Director

https://tetralogical.com<https://urldefense.com/v3/__https:/tetralogical.com/__;!!C5qS4YX3!Exyu9WjqYP1mL2R2D8DYnkgwh2Ml3vCtw9Es-00dDFc6g20Kov9NqqirkhyLd85G6PX8xnCbUcEPLbXR$>

=-=-=-=-=-=-=-=-=-=-=-=-=
Elizabeth J. Pyatt, Ph.D.
Accessibility IT Consultant
IT Accessibility
Penn State University
ejp10@psu.edu<mailto:ejp10@psu.edu>, (814) 865-0805 or
accessibility@psu.edu<mailto:accessibility@psu.edu> (main office)