Archive index

A11y Slackers Gitter Channel Archive 13th of October 2017

What fresh hell is THIS now? - Patrick Lauke
  1. zakim-robot
    @zakim-robot
    Oct 13 05:57
    [jv] Does anyone use Google Analytics?
  2. [jv] Is it accessible?
  3. zakim-robot
    @zakim-robot
    Oct 13 09:45
    [goncaloreis] Accessibility in mobile devices? How it works?
    Someone have documentation about this?
  4. Schalk Neethling
    @schalkneethling
    Oct 13 11:40
    Hey All
  5. So i am wondering. If an image does not have an alt (well, it has an empty alt), and then the image is a link to additional information. I am guessing it would be best practice to add a little descriptive text in the title attribute of the link about where it links to? To give more context, there is a header above the image and descriptive text below the image directly related to it so, perhaps that is "ok" then if both the alt is empty and there is title on the link?
  6. is no title on the link - is what I meant to say
  7. zakim-robot
    @zakim-robot
    Oct 13 12:05

    [scottohara] hey @schalkneethling. You definitely don't want to have an image with an empty alt as the only child of an <a>. For example

    <a href="https://www.my-fake-site.com">   <img src="my-img.jpg" alt=""> </a>
    

    The above is announced as:
    "link, /" in VoiceOver
    "Test / my-image link, graphic" w/ JAWS IE11 (test is the name of the folder i had this in)
    "www.my-fake-site, link" w/JAWS and NVDA in FireFox

    Your best bet would be to add a value to the image alt that would clearly announce what the purpose of the link is, for AT users.

    again, for example:

    <a href="https://www.my-fake-site.com">   <img src="my-image.jpg" alt="text here"> </a>
    

    There are minor differences in how the different screen readers announce the above, but they all consistently announce it is a link, and "text here" as the primary link text. (some also announce "graphic" or "image" with the announcement of the link and text)

  8. Schalk Neethling
    @schalkneethling
    Oct 13 12:11
    Thanks a lot @scott-morris
  9. zakim-robot
    @zakim-robot
    Oct 13 13:13
    [eric] @goncaloreis: iOS devices use VoiceOver (https://www.apple.com/accessibility/iphone/vision/) and Android devices use TalkBack (https://support.google.com/accessibility/android/answer/6007100?hl=en)
  10. [eric] both allow for full device control, provided the app or site has been built to be compliant
  11. zakim-robot
    @zakim-robot
    Oct 13 15:05
    [ghanek] lazyweb Q: what is the impact on screenreaders for text that is sentence case in the source, but formatted to appear as uppercase? This would be either Word docx or HTML source with CSS styling.
  12. zakim-robot
    @zakim-robot
    Oct 13 15:59
    [ghanek] OK, quick check with NVDA reveals it doesn't care in Word docx. Seems to even handle all caps.
  13. zakim-robot
    @zakim-robot
    Oct 13 17:19
    [vavroom] @ghanek It's what I recommend to clients who insist that they want the visual of all caps - use CSS for it. Though in general, all caps is harder to read for a lot of people, not just people with vision issues.
  14. zakim-robot
    @zakim-robot
    Oct 13 20:20
    [backwardok] anyone notice the latest update for iOS (11.0.2) started announcing elements with CSS table styling as tables?
  15. [backwardok] is the best way around this to add role="presentation" on the elements that use css tables? or is it better to file bugs against the screen reader(s) for inappropriately assigning that role?
  16. [backwardok] I did notice that VO on OSX doesn’t give CSS display tables a table role
  17. zakim-robot
    @zakim-robot
    Oct 13 22:10
    [amy.carney] Someone was asking about disability stats a few days ago. I just ran across this article today: http://www.craigabbott.co.uk/accessibility-is-not-an-edge-case
  18. zakim-robot
    @zakim-robot
    Oct 13 22:32
    [scottohara] @backwardok i have run across instances of CSS display effecting the semantics of elements before as well. for example, adding display block to <td> elements ( to make 'responsive tables' ) will negate the column/row semantics, and a table can be announced as having zero columns / rows.
  19. [backwardok] I wanna say that the previous version of iOS didn’t apply semantics to css tables
  20. [scottohara] i don't recall if it did or not. it definitely did per the display: block example I mentioned.

    firefox used to be the primary culprit for CSS display table changing the semantics of what it was used on

  21. [scottohara] but that was fixed last i checked