See also: IRC log
<scribe> scribe: AmeliaBR
<fesch> https://lists.w3.org/Archives/Public/public-svg-a11y/2016Feb/0006.html
<fesch> https://www.w3.org/WAI/PF/svg-a11y-tf/track/
<fesch> action-1705
<trackbot> action-1705 -- Richard Schwerdtfeger to Talk with at vendors about how they can present axis structures -- due 2015-08-28 -- OPEN
<trackbot> http://www.w3.org/WAI/PF/svg-a11y-tf/track/actions/1705
Fred: that can be moved for now. Jason, what about the Gamepad API figure (ACTION-1753)
Jason: I'm not sure what else I can do. I'd need to collaborate with someone who can give a full description.
Fred: For all these requests for better figures in spec, how do we want to handle it given the current state of SVG accessibility in browsers?
Jason: We at least need to make sure that all the information is in the file. But maybe we also need to link to an alternative description.
Fred: We can use described by in the SVG, but I'm not sure that's going to be read by a screen reader. These figures are inline, so there's no image alt attribute.
Jason: I think figure and figcaption have appeal.
Amelia: Many of these have visible figcaption, which is separate from an alternative description.
Jason: It would be nice to have the alternative in the SVG code itself, so it's easy to copy it into a separate file.
Fred: For the media queries figure, it doesn't have visible caption at all, it's just the text in the figure. There's a prose equivalent in the main spec text, but no specific description.
Jason: Certainly, if there is any essential syntax, it needs to be in the main document.
Amelia: We need to recognize the
difference between <img>, inline <svg>, and
<object>. For <img>, we only have alt, no detailed
structure. Could recommend authors switch to <object> if
structure is important.
... For both <object> and inline <svg>, I agree
with Jason that we should keep the alt text inside the SVG
code. We should encourage use of <desc> and
<title>, but we need to also encourage aria-labelledby
& aria-describedby to create the links to that alt text for
user agents that don't do it automatically.
Fred: But do screen readers read text inside the SVG? Not in my testing.
Amelia: They mostly do for inline code, but only with the extra ARIA roles & attributes.
Fred: OK, if that's true it makes sense.
Jason: And it means we're future-proofed for when the browsers do support SVG accessibility.
<fesch> action-2001
<trackbot> action-2001 -- Amelia Bellamy-Royds to Draft a new definition of including elements that requires title/description to be non-empty. -- due 2016-01-22 -- OPEN
<trackbot> http://www.w3.org/WAI/PF/svg-a11y-tf/track/actions/2001
Amelia: I haven't done this yet,
but I should be able to this week. Same for the other 2
actions.
... These are all clean-ups in response to Joanie's issues.
<fesch> action-2006
<trackbot> action-2006 -- Richard Schwerdtfeger to Create AAM for Graphics Module -- due 2016-01-25 -- OPEN
<trackbot> http://www.w3.org/WAI/PF/svg-a11y-tf/track/actions/2006
Rich: I was working on this yesterday. Need to fix all the build scripts within the specs. Shouldn't be too bad, since there are only three roles.
<fesch> action-1664
<trackbot> action-1664 -- Fred Esch to Change the inheritence in role=“graphicaldoc” from document to structure -- due 2016-02-02 -- PENDINGREVIEW
<trackbot> http://www.w3.org/WAI/PF/svg-a11y-tf/track/actions/1664
Rich: We also need to update the SVG spec & our SVG-AAM regarding available roles for elements to include the new roles
<scribe> ACTION: Rich to Update SVG-AAM and SVG 2 to include the new roles from the ARIA Graphics Module [recorded in http://www.w3.org/2016/02/09-svg-a11y-minutes.html#action01]
<trackbot> Created ACTION-2010 - Update svg-aam and svg 2 to include the new roles from the aria graphics module [on Richard Schwerdtfeger - due 2016-02-16].
<fesch> action-1664
<trackbot> action-1664 -- Fred Esch to Change the inheritence in role=“graphicaldoc” from document to structure -- due 2016-02-02 -- PENDINGREVIEW
<trackbot> http://www.w3.org/WAI/PF/svg-a11y-tf/track/actions/1664
<fesch> action-1746
<trackbot> action-1746 -- Fred Esch to Work on developing test cases for svg 2 accessibility & svg-aam -- due 2015-12-04 -- PENDINGREVIEW
<trackbot> http://www.w3.org/WAI/PF/svg-a11y-tf/track/actions/1746
Fred: I made the change as a
commit. Will close action 1664 if no problems
... for 1746, it's part of a bunch of actions related to the
test cases. I would like to close off the actions.
Amelia: We need to make plans to properly review these test cases, but I think you've covered your action of drafting them.
Fred: I think Chaals has an open action to review test cases. Should we create a new action for someone else?
<fesch> https://www.w3.org/wiki/SVG_Accessibility/Testing/Test_Assertions
Jason: Can you summarize what's new about these test cases?
Fred: They are all on the same wiki page, or a new page for "language-dependent" tests.
<fesch> action-2009
<trackbot> action-2009 -- Fred Esch to Create testable statements with multiple title/desc elements -- due 2016-02-09 -- PENDINGREVIEW
<trackbot> http://www.w3.org/WAI/PF/svg-a11y-tf/track/actions/2009
Fred: These are examples with
multiple title or desc. They assume that the environment is set
to prefer English.
... But I haven't got any examples in there of multiple
title/desc where some don't have any lang. I'm not sure if we
need those.
Jason: Would it take the language of the surrounding document?
Fred: I think so, based on the browser environment. But that's a good question: can you put a lang on <svg>?
Amelia: You can with xml:lang, but these are good questions & may need clarification in the SVG spec.
Fred: for the test so far, I don't make any assumptions about document language, and I'm only testing for English browser environment.
<chaals> [I do have an action to review. Please Fred also look at the stuff in http://github.com/SVG-access-W3CG - there are two repos there that have tests, and some results too]
Rich: If lang isn't specified, shouldn't it look to the nearest ancestor?
Amelia: It should, but I'm not sure it's clearly specced how that should interact with the language swapping features. Fred, do you want to review the SVG 2 spec & identify areas where behavior isn't clearly specced?
Fred: I can, I was hoping some of you might know the behavior.
Amelia: I haven't looked recently, but I don't think it's explicitly addressed.
<scribe> ACTION: Fred to review SVG 2 section on multi-lingual title/desc & identify any under-specified conditions or edge cases [recorded in http://www.w3.org/2016/02/09-svg-a11y-minutes.html#action02]
<trackbot> Created ACTION-2011 - Review svg 2 section on multi-lingual title/desc & identify any under-specified conditions or edge cases [on Fred Esch - due 2016-02-16].
Amelia: I'm sure the SVG team will appreciate. I don't think anyone's tried to implement it yet.
Fred: Joanie brought up an issue with recursive -labelledby or describedby
<fesch> Topic : Amelia clarified textPath should be treated like tspan in the AAM
Amelia: When was this? The last call?
Fred: Not really sure, just
wanted to record whether anyone had any concerns with
that.
... If not, can we log a proper resolution.
Doug: textPath should be treated like tspan
Rich: I think we already sorted that out.
Amelia: Issue came up during discussion of default roles of text elements, but this is all separate from the ARIA role="text", which has complications.
RESOLUTION: <textPath> and <tspan> should be treated equivalently, as presentational elements unless the author has added explicit alternative text or interactive features
Fred: I've been asked to make
tests for other ARIA features, such as aria-flowto. Or test
what happens if someone adds a click listener.
... does that force the element into the accessibility
tree?
Jason: I think it should, otherwise why was it added?
Amelia: That's what we agreed,
except for role=none.
... except for the obligatory role=none elements, like filter
elements, which don't represent anything on the screen.
Jason: What about aria-hidden="true"
Amelia: That to, the normal rules
apply to exclude it from the accessibility tree.
... My open actions include cleaning up this language.
Doug: Can we maybe wait until the editor has presented revised text, rather than trying to write it via a telcon?
Rich: I would follow what we do with ARIA, for cases where an element has interaction but no role, make it a group or something. Except for elements that are never mapped.
Fred: Will there be consistency issues with other specs?
Amelia: We'll try to avoid that. Point is to clarify priority between rules for including and for excluding elements.
Doug: At the SVG F2F, it was
suggested that we remove the text on focus management in SVG 2,
and instead just reference the HTML spec section.
... This is good if it improves consistency for implementation,
but bad if there aren't 1-to-1 mappings that you can clearly
interpret.
... We need to review & make sure exceptions are clearly
marked.
Rich: I wrote that section &
found a number of little inconsistencies. It will be messy if
they have to be marked one at a time.
... If they want to have a consistent spec, it should be its
own focus management module that isn't defined around HTML.
Amelia: We can create a careful list of exceptions, where SVG is different from HTML, and present that to SVG WG. Then decide whether to include that as spec or the net prose after the exceptions are integrated. Both are useful for implementers & authors.
Rich: That's re-doing a lot of work from a few years ago, taking time away from other projects.
Doug: I agree, but we need to convince people these differences are important.
Rich: I would still like to do this right, have a single over-riding module.
Doug: I don't think there will be
openness to delaying until that happens, unless we are willing
to draft that spec ourselves.
... Let's talk more about focus next week.
This is scribe.perl Revision: 1.144 of Date: 2015/11/17 08:39:34 Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: RRSAgent_Text_Format (score 1.00) Succeeded: s/GamePad figure/media queries figure/ Succeeded: s/Topic: actions pending review// Succeeded: s/If not/Fred: If not/ Found Scribe: AmeliaBR Inferring ScribeNick: AmeliaBR Present: fesch AmeliaBR richardschwerdtfeger jasonjgw shepazu Found Date: 09 Feb 2016 Guessing minutes URL: http://www.w3.org/2016/02/09-svg-a11y-minutes.html People with action items: fred rich[End of scribe.perl diagnostic output]