minutes for 27 March 2015 --- NOTE NO MEETING APRIL 3

Please note we will not have a meeting on April 3, 2015.  Next meeting will
be April 10, 2015.

http://www.w3.org/2015/03/27-svg-a11y-minutes.html


- DRAFT -
Protocols and Formats Working Group Teleconference
27 Mar 2015


See also: IRC log


Attendees
Present
Regrets
Chair
      Fred
Scribe
      cpandhi
Contents
      Topics
         1.	Chart Navigation for Sighted Users
      Summary of Action Items



<trackbot> Date: 27 March 2015


<richardschwerdtfeger> Meeting: SVG Accessibility Task Force


Chart Navigation for Sighted Users


<fesch> meeting passcode: access


<LJWatson> Meeting: SVG A11y TF telecon


<fesch> password for meeting is: access


<LJWatson> Chair: Fred Esch


<fesch> Amelia, let me know if you can get into the webshare


<richardschwerdtfeger> chair: Fred


scribenick


<richardschwerdtfeger> scribenick: cpandhi


RS Doug is not here


<richardschwerdtfeger> RS: Doug is not here


web meeting link https://apps.na.collabserv.com/meetings/join?id=2897-5157,
password access


FA: waiting for Chaals
... sharing a visual
... What we have a web gui, on the left is the specification and on the
right is the chart
... first concept to talk about is navigation order
... use arrow keys to navigate the axis, there is 1 y axis
... use the right arrow key to move across , there is a green bubble visual
focus


RS: what kind of information you have on the bubbles ?


FA: for sighted users what we are interested is only interactivity


RS: did not know if there is tool tip information


FA: for blind users we have more information but sighted users we are
interested in interactivity only
... when you have islands and you want to hop you use the shift key
... Next is box chart, we have 2 axis, x axis is the category and y axis is
the range
... depending on the statistics, there will be anomalies (outliers)
... We can cycle through the parts


Jason: is listening in


FA: use space bar to move between the parts
... next is the bar chart, x is categories and y axis the range
... when you want to move between the bars use the space bar and arrow key
to move the individual parts
... stacks are important and for blind users we give totals and a tree
structure


RS: one of the take away is you are grouping and then allowing to drill
down


FA: able to get the stack totals and individual information is important


RS: important to provide the information of the whole stack itself


FA: could provide that in the tooltip


RS: will need authoring practices real bad


FA: lets go to the cluster bar three level
... this chart s a cluster bar with 3 x axis clustered inside


RS: how do you cycle through ?


FA: You just use the arrow keys


RS: if we have an axis that have multiple levels we need roles for nested
axis


FA: for blind users this will be represented as a tree and you will get all
the information


RS: axis can have levels with multiple levels


FA: the bars are much more important when they have levels


RS: we need to make sure that any nesting and group should have roles and
can drill down


FA: With clustering it is quite normal to have multiple levels
... jump to move to relationship in a tree
... we provide relationship in a tree


RS: are you using flowto to navigate ?


FA: not yet
... the nodes coming out are visually highlighted with different colors
... use arrow keys to move to the nodes
... We give a forward path and highlight the flow from path


RS: So how do you show the differnce in color some other way ?


FA: primarily it is only by color, could add border thickness, by default
we do not
... option to highlight direction relationship
... In this graph is 3graph, with a bar char rotated by 90 dgerees and 2
pie charts
... for multiple graphs to move between the graph is important
... should be able to move between views and keep the point of regard
... Next is a Chord chart
... has 2 conceptually different sections with ribbons that connect them
... we have severities
... Next is another chord chart with countries in the outer circle and
ribbons represent surplus that connect them
... on the out side use arrow keys to move around the circle, for the
ribbons use space bar to get within
... ribbons represent relationships
... Chord charts are complex with multiple parents


Chaals: different way to represents different graphs
... this representation makes it easy to navigate


<Zakim> LJWatson, you wanted to say it essentially comes down to data,
relationships and context.


AB: When connector is not highlighted how do you which direction it is
following ?


Chaals: the idea that there is dominant parent is pointless


CP: Thank you


FA: main point across is there is lot of relations ships and data to
navigate


Chaals: what is a sensible way to do navigation that is structured
... what is the minimum information needed to describe the navigations


AB: different way to navigate, navigate by relations, need to define
multilevel navigation


FA: having a common repository is a good way to start


Chaals: use the community repository , will send an email


RS: like to do reach out to oracle to show their work


Chaals: put together a suggestion to meet in Europe, SVG group and others


Amy: i am from oracle group


RS: Amy do you charts to show ?


Amy: yes we can show some


FA: using tab key is difficult to navigate complex charts


RS: Fred you have an idea to navigate


Chaals: you have triangles inside triagles
... pick a triangle, navigate logically


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