minutes for 10 Apr 2015

https://www.w3.org/2015/04/10-svg-a11y-minutes.html



SVG A11y TF weekly telecon
10 Apr 2015


See also: IRC log


Attendees
Present
Regrets
Chair
      Fred Esch
Scribe
      fesch
Contents
      Topics
         1. Cognitive TF inter-lock
         2. Online SVG examples
         3. Navigation
      Summary of Action Items



who is on the call?


<scribe> scribe: fesch


Cognitive TF inter-lock


rs: need to set a date not next week for joint meeting with CATF


ds: is it going to be a repeating meeting?


rs: might be intermittent
... suggested Monday 12 ET


works for: doug, Fred


rs: Monday May 11 work for all?


Online SVG examples


rs: will try to set up May 11 noon ET with CATF


lw: have links


<LJWatson> https://github.com/SVG-access-W3CG/use-case-examples/tree/master



<LJWatson> https://www.w3.org/wiki/SVG_Accessibility/ExampleImages



<shepazu> https://www.w3.org/wiki/images/d/d9/SVG-UC-Airline.svg



<shepazu> https://www.w3.org/wiki/images/2/28/SVG-UC-3graphs.svg



<shepazu> https://www.w3.org/wiki/images/2/20/SVG-UC-rectrack.svg



ds: getting erros, can debug for chaals


hello chaals - talking about wiki errors with svg


amelia: if wiki is too be a problem, then could use github as published
pages


chaals: github is a pain,,, rather work with wiki...
... has svg on github


<LJWatson> @Chaals could you give me access to the repo to save forking it?


chaals: comment on examples - wonder if map to something real
... some stuff is self contained to read other examples - hard to read


<chaals> [I will give anyone who asks access to the repo]


<chaals> DS: Given that accessibilty is about communicating intent and
meaning, it is very hard to use cleaned-out examplesjust move to the things
around your "point of regard" by default


chaals: tried to provide descriptions - what you would expect in a longdesc
- comments this thing doesn't make a lot of sense ....
... also uploaded W3C flow chart diagrams... would like to get those right
... on wiki a version with what we expect to get from, for example a
stacking that makes it difficult to understand


fe: talk to me about any charts about intent


ds: we should continue on getting svg online... and will come back when we
have everything online


Navigation


rs: in SVG tiny how successful was the navigation put in?


ds: SVG tiny wasn't successful
... for mobile, some were using SVG as an UI
... Bit-Flash talked about UI and navigation - pre smart phone, limited
mobility
... reported very good directional navigation between items - two different
schemes for nav - directional 8 way nav
... you would explicitly say which element to navigate to (mapped to
direction)
... not realistic for dynamic content - hard to maintain
... : other nav - next, previous


fe: is that like tabindex?


ds: yes and no, nav next, nav previous might not be a fixed list
... with nav next, nav previous
... you set up which next and pervious
... both svg tiny nav was explicit and had to be maintained


<chaals> [+1 to using connetors as navigation paths - and allowing
arbitrary numbers of them]


ds: might be able to use connectors


jw: like invisible connectors


<fesch_> rs: by touching the screen, you could ...


<fesch_> ds: my experience of svg, some graphical elements need to be
ignored, don't want someone navigating around a bar chart, and hit a tick
mark line when they are trying to touch only the bars


<AmeliaBR_> +1 on "don't make everything focusable by default"


<fesch_> ds: favor a focusable property


<fesch_> ds: if we wanted to make things a heuristic, if it has a title it
should be focusable - and if no title, then not available to accessibility,
wider than screen readers


jw: mobile devices have two types of navigation when using AT,
... we need the invisible connector idea, valuable to be able to enumerate
what you can get to from this point, doesn't provide a next and previous
... valuable in a node and edge graph


ds: navigation from your current point?
... yes
... you can always have a next and previous...


<chaals1> [drill down, step up a level, next/prev like left/right swipe,
directional nav in maybe 8 directions, and "here is the set of places you
can get to next" ?]


rs: make use of a rotor on iOS, so you can find what is around
... we just go around in a circular manner
... so if you ran into a connector you could get information about what it
is connected to
... if we leverage that type of screen for those who can use touch... and
use circular nav


ds: suppose we have a flow chart...
... if I have neighbors left and right... and might be things on the
diagonal
... : a web rotor seems like ability to move in a random way vs next
previous


chaals: put in IRC rough navigation idea - need to move down a level and
back up... need enough pieces to do those things,
... where you have connectors, you have a logical flow,
... need to handle directional case and topology case


jw: acyclic directed graphs?


lw: worried about quite detailed - when we don't know what people want to
get out of a chart


<chaals1> [+1 to Léonie. Navigation is really a key part of analysing the
use case stuff that we need to have posted so we can think about it better]


lw: need to understand what users want out of an SVG image


<chaals1> [There are people who aren't blind and can use the mouse and
can't understand what on earth the image is about in the first place - like
me for some of the examples we have as they currently stand]


fe: we have sighted users that need to use interactivety (with keyboard)


<Zakim> LJWatson, you wanted to say we're talking about *how to get to
something, before we really know *what people need/want to discover.


jw: perception of what you want out of graphics depends on users
expectations


rs: can I write up stuff for nav


jw: likes focusable property


rs: takes action...


<richardschwerdtfeger> ACTION: Rich write up
navigational/rotor/touch/semantics with Jason [recorded in
http://www.w3.org/2015/04/10-svg-a11y-minutes.html#action01]


<trackbot> Created ACTION-1614 - Write up
navigational/rotor/touch/semantics with jason [on Richard Schwerdtfeger -
due 2015-04-17].


<chaals1> [SVG is hierarchical in its encoding but doesn't necessarily
provide the hierarchy that you actually want for navigation…]


<cpandhi> i have to leave too


ds: chaals mentioned hierarchy, svg has a hierarchy, we have toyed with the
idea of a virtual group and have that be substantive group


rssagent, draft minutes


Summary of Action Items
[NEW] ACTION: Rich write up navigational/rotor/touch/semantics with Jason
[recorded in http://www.w3.org/2015/04/10-svg-a11y-minutes.html#action01]

[End of minutes]
                                                              
                                                              
                    Regards,                     Fred         
                                                              
                   Fred Esch                                  
       Accessibility, Watson Innovations                      
    AARB Complex Visualization Working Group                  
                     Chair                                    
        W3C SVG Accessibility Task Force                      
                   IBM Watson