Re: Return of the londesc conundrum

Here is another example of providing more detailed descriptions of images from the ARIA Authoring Practices:
https://w3c.github.io/aria-practices/examples/disclosure/disclosure-image-description.html

Jon

From: Léonie Watson <lwatson@tetralogical.com>
Date: Friday, January 14, 2022 at 8:52 AM
To: Jonathan Cohn <jonathan.cohn@cambiumassessment.com>
Cc: Stephane Deschamps <w3c@nota-bene.org>, w3c-wai-ig@w3.org <w3c-wai-ig@w3.org>
Subject: Re: Return of the londesc conundrum
For images that are not part of graphical links, the solution I like is
to use a details/summary component (or custom equivalent), like this:


<details>

<summary><img src="image.png alt="Short text description here"></details>


<p>Longer description here...</p>

</details>


Léonie.


On 14/01/2022 14:28, Jonathan Cohn wrote:
> What is the suggested pattern for providing additional information about a graphic at this time? Is it Figure with fig figcaption? I would think this not ideal for more explicit descriptions of graphics. Or perhaps a link surrounding the omg often used to bring up a full size image should also have text about the image.
> Thanks!
> Jonathan Cohn
>
>
>> On Jan 14, 2022, at 7:17 AM, Léonie Watson <lwatson@tetralogical.com> wrote:
>>
>> There is almost no support for longdesc in browsers or by screen readers anymore. I haven't checked since June last year, but the results at that time are documented here:
>>
>> https://urldefense.com/v3/__https://test-cases.tink.uk/longdesc/index.html__;!!DZ3fjg!tZbXkZQEBrk6waalEydufCLSwkCRaQgiyhVt8rIpBRXRaT-CfMnP21R90onKtZM$<https://urldefense.com/v3/__https:/test-cases.tink.uk/longdesc/index.html__;!!DZ3fjg!tZbXkZQEBrk6waalEydufCLSwkCRaQgiyhVt8rIpBRXRaT-CfMnP21R90onKtZM$>
>>
>>
>> Léonie.
>>
>> On 14/01/2022 07:02, Stephane Deschamps wrote:
>>> Hello fellow accessibility people,
>>>
>>> I remember the age-long battle to keep longdesc into the spec, and then moving it to its own[1] as the main spec marked it as deprecated. It is now marked as **obsolete and non-conforming** in the HTML5.2 spec[2] so I'm at a loss as to whether one can implement it or not, considering one official recommendation versus the other.
>>>
>>> Could anyone clear that up for me please?
>>>
>>> Thanks for any input!
>>> Stephane
>>>
>>> [1] https://urldefense.com/v3/__https://www.w3.org/TR/html-longdesc/__;!!DZ3fjg!tZbXkZQEBrk6waalEydufCLSwkCRaQgiyhVt8rIpBRXRaT-CfMnP21R9FTt9FTA$<https://urldefense.com/v3/__https:/www.w3.org/TR/html-longdesc/__;!!DZ3fjg!tZbXkZQEBrk6waalEydufCLSwkCRaQgiyhVt8rIpBRXRaT-CfMnP21R9FTt9FTA$>
>>> [2] https://urldefense.com/v3/__https://html.spec.whatwg.org/multipage/obsolete.html*non-conforming-features__;Iw!!DZ3fjg!tZbXkZQEBrk6waalEydufCLSwkCRaQgiyhVt8rIpBRXRaT-CfMnP21R9xGw9iLA$<https://urldefense.com/v3/__https:/html.spec.whatwg.org/multipage/obsolete.html*non-conforming-features__;Iw!!DZ3fjg!tZbXkZQEBrk6waalEydufCLSwkCRaQgiyhVt8rIpBRXRaT-CfMnP21R9xGw9iLA$>
>>>
>> --
>> Director @TetraLogical
>> https://urldefense.com/v3/__https://tetralogical.com__;!!DZ3fjg!tZbXkZQEBrk6waalEydufCLSwkCRaQgiyhVt8rIpBRXRaT-CfMnP21R9YeNhGtY$<https://urldefense.com/v3/__https:/tetralogical.com__;!!DZ3fjg!tZbXkZQEBrk6waalEydufCLSwkCRaQgiyhVt8rIpBRXRaT-CfMnP21R9YeNhGtY$>
>>
>>
--
Director @TetraLogical
https://urldefense.com/v3/__https://tetralogical.com__;!!DZ3fjg!tZbXkZQEBrk6waalEydufCLSwkCRaQgiyhVt8rIpBRXRaT-CfMnP21R9YeNhGtY$<https://urldefense.com/v3/__https:/tetralogical.com__;!!DZ3fjg!tZbXkZQEBrk6waalEydufCLSwkCRaQgiyhVt8rIpBRXRaT-CfMnP21R9YeNhGtY$>