Unofficial log of SVG-A11y meeting, 2015 February 20

Hello all,

The W3's IRC logging / minutes system doesn't seem to be working today.
The minutes for today's meeting *should* be available at
http://www.w3.org/2015/02/20-svg-a11y-minutes.html but that is currently
giving a 404 error.

Pasted below is the raw text from IRC.

AmeliaBR


____________________________________________


<RRSAgent> logging to http://www.w3.org/2015/02/20-svg-a11y-irc

<LJWatson> zakim, this is 2742

<Zakim> ok, LJWatson; that matches WAI_SVGTF()9:00AM

<LJWatson> rrsagent, set logs world-visible

<RRSAgent> I have made the request, LJWatson

<Zakim> +Rich_Schwerdtfeger

<LJWatson> Meeting: SVG A11y TF teleconf

<richardschwerdtfeger> RRSAgent, make long public

<RRSAgent> I'm logging. I don't understand 'make long public',
richardschwerdtfeger.  Try /msg RRSAgent help

<fesch> chair: fred

<richardschwerdtfeger>
https://lists.w3.org/Archives/Public/public-svg-a11y/2015Feb/0020.html

<fesch> Topic: Abstract for SVG Accessibility Mappings from Michael Cooper

<fesch> chair: Rich

<fesch> scribenick: fesch

<Zakim> +Doug_Schepers


 <shepazu> q+

* Zakim sees shepazu on the speaker queue

<shepazu> q-

* Zakim sees no one on the speaker queue

Charles' comments:
https://lists.w3.org/Archives/Public/public-svg-a11y/2015Feb/att-0022/00-part

ScribeNick: AmeliaBR

Richard: (reading through text) This isn't really part of the WAI suite,
but I'm really less worried about the intro and more about the content.

... what about Charles' comments?

AmeliaBR:  He had a couple small points about wording.   It mostly looks
good to me, but I don't know the W3 politics of it.

Jason: It's not so much politics, as legal requirements, particularly
regarding the status section.

Richard: Charles mentions that one of the groups isn't chartered.  Is that
SVG?

Doug: It's Patents and Formats.  They've been without a charter for a year.

Richard: I don't want to hold this up waiting for them to decide on a
charter.  Does anyone have any other issues with Michael's text?

<richardschwerdtfeger> Resolution: Although there are some wording issues
that could be changed the group does not see an issue that should hold up
the publication of the SVG Accessibility API Mapping specification with
Michael’s modifications to the specification.

Leonie: That doesn't acknowledge the charter issue, which might hold things
up.

Richard: But do any of us have anything we can do about that?

... I'll send a note to W3 management, copy to the chairs and staff
contact, asking what to do.

Doug: To be clear, in the past we have let groups publish reports after the
charter has expired.  Charles was right that there are legal issues, but
the question is whether there is any unique IP material.

Richard: There isn't really.  It mostly just leverages the existing API
mapping documents.

Leonie: What is the process, in previous times when there has been
publication without an active charter?

Doug: I think publication will be approved.  But this isn't an ordinary
working draft update.  As first publication, it is a patent disclosure
opportunity; it won't be the last, but...  There is a proposed policy that
groups should not be allowed to publish without a charter, but I don't know
the status of that resolution.

<richardschwerdtfeger> Topic: Change proposal for a rewrite of title and
desc in the SVG2 specification

ACTION richard: send email regarding PF Group charter status and how it
affects publication of API mappings

* trackbot is creating a new ACTION.

* RRSAgent records action 1

<trackbot> Created ACTION-1588 - Send email regarding pf group charter
status and how it affects publication of api mappings [on Richard
Schwerdtfeger - due 2015-02-27].

<richardschwerdtfeger>
https://lists.w3.org/Archives/Public/public-svg-a11y/2015Feb/0018.html

<richardschwerdtfeger>
https://lists.w3.org/Archives/Public/public-svg-a11y/2015Feb/att-0023/00-part

Richard: (discussing Charles' comments) He asks about alt -- we don't have
an alt attribute do we?

<shepazu> q+

* Zakim sees shepazu on the speaker queue

Doug: I think he's saying we should look at what HTML says about the alt
attribute, and make sure we have an equivalent.

... alt doesn't create a tooltip, unlike HTML title attribute

AmeliaBR: we have title and desc to do that

Jason: And this would all be available as the accessible name / description
correct?

<richardschwerdtfeger> q?

* Zakim sees shepazu on the speaker queue

Richard:  Yes, all the information about the name computation have been
pulled out and the role of the elements clarified.

<richardschwerdtfeger> ack shepazu

* Zakim sees no one on the speaker queue

Jason: The one thing where we would want to be specific for SVG is how to
create good alt text.  The guidance for HTML is for the image as a single
element, but SVG has descriptions/titles for each part.  But that's more
for an authoring guidance document.

Doug: I would like to emphasize that SVG is not always an "image" -- it can
be mostly text, after all.  On a side note, I discovered recently that
VoiceOver reads out SVG content even when the SVG is embedded as an HTML
<img>.

... On the main topic, I'm not super enthusiastic about your text.  Is
written more as authoring guidance, I would like to see tighter wording
about what a user agent MUST do, and what an author MUST or SHOULD do.

Richard: On the strict requirements, I was a little unsure.  I would like
to say that a <title> element should be a tooltip, but I'm not sure whether
we can just order browsers to do that.

Doug: At the least it should be MAY, I would like to make it SHOULD.  We
can also add an informative note, mentioning that existing implementations
treat the text content of <title> as tooltip content, and future user
agents should follow.

... For the actual normative requirements, we would say that agents SHOULD
make <title> text available as tooltip or similar, and that they SHOULD NOT
use <desc> content as a tooltip.

Richard: OK.  I was working from what we had.

Doug: What we have is not great.  Do you want me to take a stab at
re-writing?

Richard: I'm fine either way.

Doug: OK.  It's possible that when I take it to the SVG WG they'll tell me
I'm completely wrong, but I would really like to make clearer conformance
requirements -- in many parts of the SVG specs.

Richard: What we could do is separate out a section that is clearly
guidance for authors, and the rest is specifics for the user agent.

Doug: Maybe, so long as it's clear and not cluttered.  But either way, this
is an improvement -- I just think it can still get better.

Richard: One other thing from Charles' comments, he says that "describedby"
can only point to text, and that's not true.  It can point to complex
content, and the normal rules apply for accessing the description from that.

Doug: One thing I don't like about the original text, that you've copied
over, is that it talks about <desc> and <title>.  It should be the
opposite: <title> and <desc>.  Not only is <title> more important, but it
MUST be the first child.

... that was a performance issue from the browsers.  There is only one
title, and it must be the first child to be recognized.

Richard: But didn't we discover there can now be multiple titles?

AmeliaBR: Only as multi-lingual alternatives, with a language attribute.

Doug: Does that require a <switch>?

AmeliaBR: Now, the switch system didn't work with title requirements, so
it's a separate system.

Richard: One thing to note is that most content will be mapped as generic
containers, and there is currently no expectation that all of the elements
in the document will have <titles> or <desc>.

Doug: That's appropriate.  We don't want to advise that authors use <title>
for every element; that would just make a mess; separate authoring guidance
could suggest how to make good titles and such.

Richard: What about the appendix?  Should I just delete that?

Doug: Or, given that it is such a long spec, it could still be useful to
have a summary of all the related content.

Leonie: Linking back.

AmeliaBR: +1 to having an accessibility index linking to the specific
sections

ACTION richard: Update the Accessibility Appendix in the main SVG specs, to
link back to the relevant content

* trackbot is creating a new ACTION.

* RRSAgent records action 2

<trackbot> Created ACTION-1589 - Update the accessibility appendix in the
main svg specs, to link back to the relevant content [on Richard
Schwerdtfeger - due 2015-02-27].

<richardschwerdtfeger> Topic: CSUN Tuesday evening

Fred: Our current plan is to meet up sometime after 5pm

Richard: I have to speak at the 30th anniversary event, and I think thats
somewhat early

* shepazu upset he wasn't asked to speak at the gala :P

Leonie: Isn't that Thursday?

Richard:  OK then, I think that Tuesday is still open.

(more discussion on dinner and dietary requirement.  Vegetarians don't eat
fish...)

<richardschwerdtfeger>
http://www.10best.com/destinations/california/san-diego/restaurants/mexican/

* AmeliaBR tangent from shepazu's insistence that fish are not vegetables:
http://www.slate.com/articles/life/food/2010/04/consider_the_oyster.html

Richard: A reminder to all about the panel session, which I think is
Thursday morning (9am)

... And I think Doug is doing some specific talks about graphics

Fred: at 8, 9, and 10am.

...Friday: at 8 on Describler, at 9 on color blind issues, at 10 on charts
for the blind

Richard:  How are you promoting this?  Is it through IBM?  On the main
schedule?  I can post / tweet the IBM schedule once its available.

Doug: We have a nice block of accessible graphics talks there Friday
morning, we should try to promote it ourselves.

Leonie: We're out of time to talk about the scope / user requirements
document.

Fred:  Everyone can keep contributing to the wiki.  Some sections are still
looking lonely.

Jason: I think after CSUN will have some more stuff.

Leonie: I'm really interested in this stuff (requirements and authoring
guidance), so I can take a lead on this if people would like.

Jason: I'm sure we'll have lots of relevant discussion at CSUN, so if you
can make sure that all those points migrate to the wiki afterwards, that
would be good.

Doug:  I'm really hoping that the panel next week will turn out to be a
good chance for feedback from really experienced accessibility people.  We
really want to promote that.

zakim, list participants

<Zakim> As of this point the attendees have been Fred_Esch, [IPcaller],
+1.609.759.aaaa, Rich_Schwerdtfeger, Doug_Schepers

<richardschwerdtfeger> zakim, bye


 <Zakim> leaving.  As of this point the attendees were Fred_Esch,
[IPcaller], +1.609.759.aaaa, Rich_Schwerdtfeger, Doug_Schepers

zakim, [IPcaller] is me

<richardschwerdtfeger> RRSAgent, make minutes

<RRSAgent> I have made the request to generate
http://www.w3.org/2015/02/20-svg-a11y-minutes.html richardschwerdtfeger

RRSAgent, make log public

<RRSAgent> I have made the request, AmeliaBR

RRSAgent, make minutes

<RRSAgent> I have made the request to generate
http://www.w3.org/2015/02/20-svg-a11y-minutes.html AmeliaBR

(FYI to all, I'm still getting 404 errors on that link.  If anyone can
access it, please email to the list)