Meeting minutes for 11/21 meeting (just in time for the 11/28 one)

-        Action items have been updated, see
https://www.w3.org/community/auto-wcag/wiki/Action_items#Open_actions_from_a
uto-wcag_Community_telecons

 

Meeting notes for auto-WCAG meeting

Nov. 21, 2014

Scribe: Birkir Gunnarsson

Attendants (according to the irc channel):

Wilco, Annika, Birkir, Katie, Matthieu, Frank, sharon 

action items:

https://www.w3.org/community/auto-wcag/wiki/Action_items

Updates on the Wiki:

- Annika has fixed the tables, now table headers in first column are headers
for rows.

- We still have not addressed the issue of using CSS that does not break SC
1.4.1.

Annika is going to contact Michael Cooper

- Kamyar is doing research on 3.2.3, should have updates by next week.

- Birkir: Twitter account has been built and is now active .

Annika, the link to it should now be on our homepage (Birkir does not find
it on 11/27, but perhaps it is not accessible).

Action item: Everybody follow autowcag and retweet if interested.

Action item: Auto-WCAG (and the auto-WCAG Twitter account) need a logo

Birkir/Wilco: follow up on that.

 

review of 1.4.1

https://www.w3.org/community/auto-wcag/wiki/SC1-4-1-a

(notes written with all the understanding and grace of a man who has been
blind for 32 years .. appologies if something here is misconstrued).

 

Color contrast:

Problem:

What when an image is used as background for both links and surrounding
text? Hard to determine color contrast accurately.

What about computed styles?

Should we go all the way up to the root to determine?

Once you get up to the html tag and there are no styles on it, we can assume
page is using the 

browser default.

What do we do if there is no declared background color?

Are there any technologies that compute color on rendered pages?

 

Another problem: using background color to distinguish link from surrounding
text:

Selectors can not necessarily figure that out accurately.

In our tests, we have to explicitly assume that we test foreground color
change only, not background.

 

- Test should cover focus, (i.e. move focus on the link and recalculate), in
step 3 of the test.

 

review 4.1.2 - frames

https://www.w3.org/community/auto-wcag/wiki/SC4-1-2-frame

thoughts:

- Are frames considered 1.1.1 - they are non-text content (they are also
frequently used for 1.2, 

multimedia content is frequently displayed in frames)?

- We need to make it clear in language of test that by "frame" we refer to
both frames and 

iframes.

- If frame title is missing or empty, then fail.

- What about frames or iframes that are hidden (used by server to track user
traffic)? In general 

This is hard to judge accurately since often content can be hidden on a page
when it loads initially (such as inactive tabs), but could at any time be
revealed as a result of a user action. We should require that all frames
have titles but when communicating failure to user, we need to indicate the
possibility that the fram is not visible at the time, ... how best to do
that?

related Action item for Wilco, update our naming conventions for test cases
that cover multiple success criteria to reflect which ones are relevant.

 

Avoid duplication, provide documentation (this is a rhyming sensation)!

 

final thoughts:

Frank - 1.1.1 is complicated!

Birkir - 2.4.4 tests being worked on, should be ready for review either on
Dec 4th or 11th.

11/27 is Thanksgiving day in the U.S., people will likely be too busy
cooking turkey to join this meeting. Once they quit cold turkey they can be
back with us.

 

 


Birkir Gunnarsson
Sr. Accessibility Consultant
 <http://www.deque.com/> Deque Systems Logo

	

2121 Cooperative Way | Suite 210 | Herndon, VA 20171-5344
tel (919) 607-2753 


 <http://www.deque.com/> website |  <mailto:birkir.gunnarsson@deque.com>
email 


 <https://twitter.com/dequesystems> Deque on Twitter
<http://www.deque.com/blog> Deque's Blog
<https://www.facebook.com/dequesystems> Deque on Facebook
<http://www.linkedin.com/company/358829?trk=tyah> Deque on LinkedIn


Skype Username:  wildebrew77 


Confidentiality Note: This email may contain confidential and/or private
information. If you received this email in error please delete and notify
sender.