The Cognitive Accessibility Task Force Teleconference -- 11 May 2015

<Lisa_Seeman> agenda: this

<Lisa_Seeman> hi, are you joing us on webex or zakim?

<Lisa_Seeman> we are trying to switch...

<Lisa_Seeman> https://mit.webex.com/mit/j.php?MTID=m84fb2a12467c1fb271601e13b78022b3

<Lisa_Seeman> MicahelC we can NOT strt the webex meeting without you!

<Lisa_Seeman> https://mit.webex.com/mit/j.php?MTID=m84fb2a12467c1fb271601e13b78022b3

<Lisa_Seeman> can people joint the webex

<Lisa_Seeman> US Toll Number: +1-617-324-0000

<Lisa_Seeman> Access code:649 368 070

<Lisa_Seeman> Mobile Auto Dial:+1-617-324-0000,,,649368070#

<Lisa_Seeman> or go to https://mit.webex.com/mit/j.php?MTID=m84fb2a12467c1fb271601e13b78022b3

<richardschwerdtfeger> https://www.w3.org/WAI/PF/cognitive-a11y-tf/wiki/Svg_comments

<Lisa_Seeman> next itm

<Lisa_Seeman> next item

<Lisa_Seeman> Doug can you join us on webex

<Lisa_Seeman> next item

<chaals> oh. webex

<Lisa_Seeman> https://mit.webex.com/mit/j.php?MTID=m84fb2a12467c1fb271601e13b78022b3

<ayelet_s> I could

<ayelet_s> pleasure

<Lisa_Seeman> scribe: ayelet_s

<Lisa_Seeman> challs can you callin

<Lisa_Seeman> US Toll Number: +1-617-324-0000

<Lisa_Seeman> Access code:649 368 070

<Lisa_Seeman> Mobile Auto Dial:+1-617-324-0000,,,649368070#

<Lisa_Seeman> -------------------------------------------------------

For assistance

<Lisa_Seeman> -------------------------------------------------------

1. Go to https://mit.webex.com/mit/mc

<Lisa_Seeman> 2. On the left navigation bar, click "Support".

<Lisa_Seeman> chaals - can you call in?

<richardschwerdtfeger> Rich is here

<Lisa_Seeman> Lisa is here

Michael: need different way to take attendance

<Chaohai_Ding> Chaohai is here

<Lisa_Seeman> Chaalls -you can goto https://mit.webex.com/mit/j.php?MTID=m84fb2a12467c1fb271601e13b78022b3

<Lisa_Seeman> online

<Lisa_Seeman> no need to call

<Lisa_Seeman> Also Janina is fining a url were you can enter your number and webex will call you in

<MichaelC> https://mit.webex.com/mit/j.php?MTID=mb10f72a1b8d5c8972343dc644e3c930c

<Lisa_Seeman> https://www.w3.org/WAI/PF/cognitive-a11y-tf/wiki/Svg_comments

<Lisa_Seeman> chaalls did you try https://mit.webex.com/mit/j.php?MTID=mb10f72a1b8d5c8972343dc644e3c930c

<Lisa_Seeman> (diffrent link) it will call you

<Mike_Pluke> You mentioned a link in IRC - only just joined IRC so can't see link. Can someone paste it in IRC again?

<MichaelC> https://mit.webex.com/mit/j.php?MTID=mb10f72a1b8d5c8972343dc644e3c930c

<Lisa_Seeman> I can here u

<Lisa_Seeman> yup

did fesch introduce the comennts SVG is looking for?

<fesch> rich introduced SVG comments

thanks

rich: want to know what we need in SVG to support cognitive disabilities

<chaals> [I note that at a minimum, we need rational navigation, and the ability to identify things from each other…]

<chaals> [identify visually, I mean. Authoring requirement as much as anything]

<Lisa_Seeman> https://www.w3.org/WAI/PF/cognitive-a11y-tf/wiki/Svg_comments

fesch: is simplification a wanted requirement?

lisa: for some users the ability to simplify is useful
... there is a shift from the encoding to the content
... so for example, use symbols to aid understanding
... the ability to highlight part of the visual content will be helpful
... tags (possibly aria) for numbers and non literal content, with which to provide alternative content. will translate to visual content as well

<Lisa_Seeman> ok

<Lisa_Seeman> DOug: authering tips is outside scope

<Lisa_Seeman> Doug: we are interested in wht sematics are need

<chaals> lisa: When there is text or speech describing things, be able to associate it with the part(s) of the image being described…

<Lisa_Seeman> rich: what would the user experence be

Fred: this is the challenge - "clear symbols" aren't defined accurately

who said that ^

<richardschwerdtfeger> Plumbing Potential Features:

<richardschwerdtfeger> * Media Query feature that says to expose features of a given importance level or higher

<richardschwerdtfeger> * Need the ability to group sections of a diagram together with a defined keyboard navigation order

<richardschwerdtfeger> * Ability to highlight content with their associated visual descriptions

<richardschwerdtfeger> * Ability to annotate the importance level

lisa: the current tags might be the wrong ones for the cognitive group - we want non literal, nunmbers and importance

<chaals> LS: People want to be able to request fewer, smaller icons. Also to be able to get information about things they are focusing their attention on.

<inserted> scribe: chaals

LS: e.g. using something like aria-described - but as I mentioned before, the content might be different to what woulld be requested by a non-visual user

DS: Best thing is to decide what we are trying to accomplish - break down our work into different goals.

<LJWatson> +1 to identifying goals.

… one thing will be identifying use cases

[+1 to identifying goals]

… some aspects of use cases will lead to requirements for users that are fulfilled by authoring practice, some by adding features to SVG or ARIA, some will crossover

… Some will be requirements on user agents.

… It is relatively easy to identify the features needed by a screenreader

[For the record I don't think it is that easy with a screenreader either, but doug's point is still valid]

… There are different kinds of requirement, and we need to be able to work out how to handle the range of them - e.g. signalling user requirements and use profiles

LS: We have been working on doing this client-side to pull out the things people want without giving away their privacy / security

<richardschwerdtfeger> Plumbing Potential Features:

<richardschwerdtfeger> Media Query feature that says to expose features of a given importance level or higher

<richardschwerdtfeger> Need the ability to group sections of a diagram together with a defined keyboard navigation order

[+1 to making sure you don't identify how easy it is to trick you in a certain way…]

<richardschwerdtfeger> Ability to highlight content with their associated visual descriptions

<richardschwerdtfeger> Ability to annotate the importance level

<richardschwerdtfeger> Need alternative, alternative text for people with cognitive impairments

<richardschwerdtfeger> We need an aria-nonliteral attribute (takes a string). See the example above. (Note: Lisa, what do you want the user agent or author to do with this. e.g. if there was a media query that stated that non-literals are used instead ... render it some how)

<richardschwerdtfeger> We need less icons but make them bigger. ...?

<richardschwerdtfeger> We need to be able to pull in information about a picture that is pulled in from somewhere else. (To Do for Lisa to expand on this)

<richardschwerdtfeger> Ability to tie colors to semantic meaning?

DS: So how do we want to approach this stuff. Do we need a longer conversation?

LS: I heard two things. One, identify the goals. 2. We need to understand the user experience and what it needs to be.

… then we can get into the nitty gritty of how we might try to achieve these things.

RS: Two things I see here. We need a vehicle for the user to specify their needs.

… e.g. like user context API

… Sounds like we also need additional attributes in content to help the user agent or application author match content to user requirements.

… We also need to know what the user agent needs to do.

… we can add alternate content and similar metadata of various kinds.

… We will need some of these things for SVG drawings. They are not necessarily important for assistive technology using accessibility APIs but it does impact the accessibility of the user experience.

<Lisa_Seeman> LS: I heard two things. One, identify the goals. 2. We need to understand the user experience and what it needs to be. 3. how the user comuntes their need

… We need a way for the user to get the "correct" content.

s/comnutes/communicates

<Lisa_Seeman> One, identify the goals. 2. We need to understand the user experience and what it needs to be. 3. idetfy the sematics we need, 4 how the user comuntes their needs to the application or user agent

<LJWatson> Chaals: What we need is a statement of the problems people face. A good way to do this is to gather as many examples for as many different disabilities as we can.

<LJWatson> Chaals: Before we start assigning changes to UA, ARIA or SVG, we should understand those questions at a basic level at the least.

<Lisa_Seeman> https://w3c.github.io/wcag/coga/user-research.html

<Lisa_Seeman> https://rawgit.com/w3c/coga/master/techniques/index.html

<Lisa_Seeman> https://www.w3.org/WAI/PF/cognitive-a11y-tf/wiki/Syntax_for_adaptable_links_and_buttons

<LJWatson> Chaals: If you have an attribute that says this content doesn't contain numbers, I think that's the wrong way around. We should try to identify *when content contains numbers. There's a lower risk of false positives, and it's easier to accomplish.

<LJWatson> LS: Will write up some conclusions from this meeting and circulate.

-> https://www.w3.org/wiki/SVG_Accessibility#Images_for_playing_with_to_determine_use_cases_and_requirements A growing set of SVG images to consider …

<LJWatson> LS: My concern is that with so many important deadlines coming up, how do we prioritise?

[needs a lot of things like interactive interfaces and applications to be added]

<LJWatson> LS: Could we put CogA people in SVG TF or vice versa perhaps? Thoughts?

[I suggest we prioritise mostly by focusing on the stuff that people do the work to move forward…]

<LJWatson> RS: Not sure of SVG 2 timeline, but would like to get something up and running within the next 12 months. We have work to do!

[chaals leaves for real]

<LJWatson> RS: We (the CogA TF) need to identify the attributes that match the needs of users, and develop guidelines on how to use them properly.

<LJWatson> RS: Also need a vehicle to convey to the UA how to pick the right experience - it's a context issue.

<janina> Need to leave now, sorry!

<Tony_Doran> I need to drop off sorry

<ayelet_s> yes\

<Lisa_Seeman> scribe:ayelet_s

<Lisa_Seeman> https://w3c.github.io/wcag/coga/user-research.html

Fred: we need actual research on user cases

Lisa: we have that in the user cases, but they are not specific to visual user cases

<Lisa_Seeman> https://rawgit.com/w3c/coga/master/techniques/index.html

,,, worth reading them before deciding to make more user cases

scribe: we are writing JSON object notations for adapting the content for different user group
... will be deployed quicker than GPII

Fred: do we have user cases about charts or graphics

LS: no, but we have techniques

Fred: we need to extract ones for SVG

<Mike_Pluke> I've lost the Webex audi after losing the connection and rejoining - have we nearly finished?

<Lisa_Seeman> https://www.w3.org/WAI/PF/cognitive-a11y-tf/wiki/Easy_Personalization

LS: what we did so far. is a gross approach but could become more complex

*link to what we did so far

Lisa: will write E-mail in conclusion. Will also go back to task force and discuss how to prioritise this

Fred: how could we(SVG) validate the techniques you propose?

LS: easiest way would be to bring it back to our task force, as we have members working with people with disabilities
... Thanks for all, and sorry for the problems with the new technology

- DRAFT -

The Cognitive Accessibility Task Force Teleconference

11 May 2015

Attendees

Contents

For assistance

1. Go to https://mit.webex.com/mit/mc

Summary of Action Items

Scribe.perl diagnostic output