CN: small number of requirements. need to identify testable statements. should be able to find proper implementations of longdesc and get to PR.
...need to identify tools that demonstrate support, should be the case with multiple AT and plugins.
...browsers currently do a terrible job.
...need to describe what happens when you have a base url specified in the head of the document. this could be a problem.
...we need people who can write and run tests. Need a setup that supports longdesc in some way.
DM: I have windows with JAWS 14, and a mac
...and NVDA
LW: I can test with NVDA as well
JS: I can do orca on linux
LW: I can do JAWS 15 as well
CN: regarding replies to comments made, I will prepare some drafts and share them with the group.
JS: There is a side conversation going on regarding Math ML. I'm presuming we will say that we support people using accessible markup whenever possible. Just some context on that side convo.
<chaals> MS: Last week we replied to HTML group and came up with a summary of what we need to look at in terms of interoperability (or lack of).
<chaals> ... that will be out "Real Soon" (hopefully this week).
<chaals> ... For canvas, Rich and I have agreed to set aside some time going through the 2d API and looking for issues in that.
<chaals> ... Would be good to have a sub-team on HTML5 testing since we will need to start working closely on accessibilty-specific tests.
<chaals> CN: Think that makes sense.
<chaals> MS: Could be a generic "testing" sub-team.
<chaals> JB: Sounds good.
CN: relatively little implementation follow-up of earlier SVG accessibility work.
...expects to look at more ARIA properties for improving the a11y of SVG, specifically visualizations of data.
JB: in terms of existing SVG stuff, SVG 1.1 still has some ways to go RE implementations. some of the SVG 2.0 a11y work that is already underway is something we wouldn't want to conflict with. They are still sorting out priorities and timelines.
...think there is great possibility in a11y of data visualizations.
DM: more on Data viz and ontologies?
JB: data org charts, use connectors, no accessibility access to these relationships
...important area for exploration. hoping that the new community group is aware of the existing work on accessibility being done in SVG 2; maybe we can get more energy on that.
CN: the goal is not to augment the work that has already been done.
JS: Rich is actively engaged with the SVG WG and an editor on the SVG 2.0 document and is working directly with them to get ARIA support.
...ARIA 1.1 should also start bridging any gaps in SVG a11y.
...also hoping to get requirements in place for ARIA 2.0 which will focus on graphic accessibility. The new CG could develop use cases. Would be very helpful.
...ARIA 2.0 may not be a monolithic spec, but a modularized method similar to CSS
CN: we will do our best to keep coordination and communication between the groups open.
JS: spent a fair amount of time last week asking people to do a detailed review. Do we have the right amount? will it stay stable? Talked about taking this up with the WAI CG group.
CN: i think longdesc will have an impact on this as well.
...need to make sure that nothing there is bad advice.
JS: i.e. logo, perhaps we could do better, such as indicating the category of image along with descriptions.
DM: how many people think it is OK to put the word logo in alt text?
JS: i think its critical.
CN: have we gotten rid of the things we wanted out of the spec, what do we need further discussion on and what is missing.
...that sets us up to do some useful work.
JS: agreed
CN: think we should take this up in the media sub-team.
JS: that is on my list of todos and I can send out an email to see what people's availability is.
...there are a couple of reasons to bring the media sub team together.
CN: the CfC in the TF has expired and it has passed. I have not asked for review by PF and HTML WG yet. Will do when I get network access back
no sub team reports. Media will be active, Bug Triage is quiet.
DM: anyone know if SEO indexes details element?
CN: search engines always cautious about spam. elements like details should be perceived as relatively safe and indexed.