<scribe> scribe: clapierre

<pkra> +1 for Pluto.

tzviya, we sent our note to PF for ARIA-describedat

…, Suzan Taylor sent an email back in 2011 and we want to make sure there is no contradictions with what we sent

Deborah, thinks that that it doesn't

tzviya, thinks that maybe describtedat wasn't even mentioned

deborah, no one will say they will use this because no User Agents / Browsers support it.

…, we can push back that the publishers doesn't have to say we will use it until there is UA/Browser support

tzviya, will reach out to Suzan to make sure there is nothing we should be concerned about and we are not contradicting ourselves with that previous email with our new note to PF.

<tzviya> 1. Discuss the ARIA Role attribute ‘math’ with invited guest Peter Krautzberger

<tzviya> 2. Discuss the ARIA-describedat document we presented to PF and ARIA, next steps…

<tzviya> 3. Discuss the continuation of the work for the A11Y DPUB Note

<tzviya> 4. Discuss the changing of the time for this meeting to allow for broader participation.

math and ARIA Role

<pkra> http://w3c.github.io/aria/aria/aria.html#math

Peter, current aria spec is not great with respect to math.

…, only a section concerning polyfill. Can this be changed?

…, more and more tools that will render MathML, SVG, etc…

…, browser venders are not implementing MathML

…, IE / Chrome:Blink not on their radar to implement MathML

<pkra> MathML?

clapierre, Benetech is writing a Web Based database to determine what is available for a given set of criteria (ie. Windows / Jaws what works with that.

deborah, image renders when MathML is not supported otherwise display the MathML. But this is currently a problem.

Peter, talked to Sandars, you really have to fall back to binary images without markup.

…, we can produce SVG from the mathML, change for us at MathJax we will have in the next release HTML and CSS instead of SVG for displaying math equations from mathML.

…, math player with IE back in the day was easy you could detect that and switch… much more complicated now.

…, hide mathML in DOM with aria-hidden

…, but this is bad practice because you break the visual with the accessible

…, mathJAX feels this should be overcome… longterm it seems feasible that something can expose the mathML

…, in browser they are not able to style math elements which is a big problem.

tzviya, next steps: contact PF and discuss aria-role math will be..

…, aria 1.1 mid august will be finalized

…, aria 2 a year away

peter, AT seems to have good means to us mathML

…, thanks to Benetech and work Peter helped with.

tzviya, Jone Marie suggested more role attributes with mathML in aria-role

tzviya, what would you like to see happen Peter?

peter, have non mathml markup that AT can identify as mathML markup

…, math roles only work on mathML

…, seems like the obvious step would be to expose the different formats (eg. LATex )

…, first step is to expose the mathML

…, we can do hacks but we want something more mainstream

tzviya, the polyfill section does deal with this does it not?

<tzviya> http://rawgit.com/w3c/aria/master/core-aam/core-aam.html#mapping_role_table

peter, that is just then going to show the alt-text

…, the hack is just to use data attributes

…, maybe that would work but that is not how it should work.

tzviya, link provided on how AT should handle this role for math.

peter, you find a huge # of SPANS of a page rendered from mathML

tzviya, but wouldn't you reproduce the mathML in ARIA?

…, Peter write up a summary of the problem in detail

…, explain on how this works today

…, explain the mapping of the Accessibility Tree sees

…, if you have a solution then propose it, and if you are not sure of the solution then you can ask as well.

peter, ok I will do that.

deborah, we all agree there is a problem

… a 3 sentence problem statement "elevator summary" and even if you don't have a solution… if you do great

tzviya, you can say that reproducing mathML in ARIA works but there must be a better solution to the problem and ask for help.

peter, danger to mathML if there is rendering that is just as good

…, there is really nothing moving forward for mathML

deborah, we should be on the record that Browser manufactures need to implement mathML standard, which will solve our accessibility issues, instead of ask ARIA and others to add new standards to address the inaccessibility of mathML because it isn't being implemented by most browsers.

meeting time

tzviya, someone from Sage publishing wants to join, Thierry wanted to come but the time does not work.

deborah, we don't want to reschedule which will affect the core people.

mia, any earlier would be a problem for me.

tzviya, maybe send a request to DPUB to see who else would be interested and find out why they can't connect currently

…, most people are on vacation during the summer

…, sounds like we should keep this meeting at this time for now

deborah, may have a conflict with a stand-up, but will ask to move that.

charles, will be on vacation first week of August

tzviya, can someone from Benetech join from BookShare to talk about how they foresee using describedAt

charles, yes I will talk to my colleagues and find some representative. Maybe Anh Bui would be ideal. Or Robin.

deborah, we have gotten distracted by this aria-describedAt and once that is put to bed we will shift focus to our a11y Note for the W3C

tzviya, remember we are proposing this Note be complete by year end.

charles, yup… sounds good.

rssagent, draft minutes

This is scribe.perl Revision: 1.140  of Date: 2014-11-06 18:16:30  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: s/Theiry/Thierry/
Found Scribe: clapierre
Inferring ScribeNick: clapierre
Present: tzviya_siegman Mia clapierre Peter Krautzberger dkaplan

WARNING: No meeting title found!
You should specify the meeting title like this:
<dbooth> Meeting: Weekly Baking Club Meeting


WARNING: No meeting chair found!
You should specify the meeting chair like this:
<dbooth> Chair: dbooth

Got date from IRC log name: 17 Jul 2015
Guessing minutes URL: http://www.w3.org/2015/07/17-DPUB-minutes.html
People with action items: 

WARNING: Input appears to use implicit continuation lines.
You may need the "-implicitContinuations" option.