W3C home > Mailing lists > Public > public-html-a11y@w3.org > August 2013

RE: Testing longdesc...

From: Léonie Watson <tink@tink.co.uk>
Date: Wed, 21 Aug 2013 22:13:02 +0100
To: "'Charles McCathie Nevile'" <chaals@yandex-team.ru>, <public-html-a11y@w3.org>
Message-ID: <005d01ce9eb3$38e8f990$aabaecb0$@tink.co.uk>
Chaals McCathie Nevile wrote:
"These can each be used to test 3 different requirements: whether the long description is available to a user, whether it is exposed to an accessibilty API if the user agent connects to one, and whether the user can discover that a long description is available."

Some initial screen reader test results with NVDA 2013.1 and Jaws 14, on Firefox 23 and Internet Explorer 10. https://dl.dropboxusercontent.com/u/2805347/Image-Description-screen-reader-test-results_2013-08.xlsx

The last test (invalid longdesc (plain text)) wasn't available. I used images with the following longdesc variations to test:

Invalid longdesc (plain text): longdesc="12345 
Invalid longdesc (empty): longdesc="

Léonie.
-----Original Message-----
From: Charles McCathie Nevile [mailto:chaals@yandex-team.ru] 
Sent: 13 August 2013 15:10
To: public-html-a11y@w3.org
Subject: Testing longdesc...

Hi,

I wrote up some tests that I propose we consider for the longdesc spec.  
They are basically manual tests.

Most of them are different combinations of longdesc types - internal "hash" reference, external url, data url, URL surrounded by whitespace - in different conditions - missing image, image is a data url, there is a base element, and so on.

These can each be used to test 3 different requirements: whether the long description is available to a user, whether it is exposed to an accessibilty API if the user agent connects to one, and whether the user can discover that a long description is available.

There are a couple of others I added to test specific things like what happens when the longdesc URL is changed by a script.

There is probably something missing, and there may be errors in, or improvements to be made to the tests I propose.

They are on github so you can propose changes as pull requests, if you're a github wiz - or by email if you like. We need a process for formally agreeing to adopt any test we want to use, and these are just personal input into that process...

[1] http://github.com/chaals/longdesc-tests

--
Charles McCathie Nevile - Consultant (web standards) CTO Office, Yandex
       chaals@yandex-team.ru         Find more at http://yandex.com
Received on Wednesday, 21 August 2013 21:13:33 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 19:56:27 UTC