Accessible Diagrams

classic Classic list List threaded Threaded
14 messages Options
Reply | Threaded
Open this post in threaded view
|

Accessible Diagrams

chatai
Hello Everyone,

My name is Cagatay Goncu, known as Chatai and I am studying PhD at
Clayton School of Information Technology, Monash University,
Australia. My research is on accessible diagrams. With this research,
we are investigating to develop techniques for presenting diagrams in
different media to blind people in different layouts and modalities
such as tactile, audio, and tactile/audio.

I am currently working on a tool that gets data from a text input file
and creates a tactile bar and pie chart as an SVG file. We are
planning to extend this tool to other types of diagrams such as
network diagrams, schematic diagrams, and maps with other types of
modalities. And to achieve this objective we will probably define a
structured SVG format for these kinds of graphics.

Since you all have been working on similar topics, I am wondering
whether we can work together?

Cheers,

Chatai
_______________________________________________
dev-accessibility mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-accessibility
Reply | Threaded
Open this post in threaded view
|

Re: Accessible Diagrams

Alexander Surkov
Hello, Chatai.

We always are open for collaboration or try to be open at least :)

In two words all we do is we expose pages rendered by Gecko to AT by
AT API (like MSAA, ATK, IAccessible2). Currently in Gecko we support
HTML, XUL and XForms. In our plans is to support SVG as well. But
still we didn't final approach to do this (or it's unknown with me).
Sure we search a way how to expose SVG markup via AT API. If you keep
in mind something similar then we may be helpful for each other.

Alexander.


On Apr 16, 9:35 am, chatai <[hidden email]> wrote:

> Hello Everyone,
>
> My name is Cagatay Goncu, known as Chatai and I am studying PhD at
> Clayton School of Information Technology, Monash University,
> Australia. My research is on accessible diagrams. With this research,
> we are investigating to develop techniques for presenting diagrams in
> different media to blind people in different layouts and modalities
> such as tactile, audio, and tactile/audio.
>
> I am currently working on a tool that gets data from a text input file
> and creates a tactile bar and pie chart as an SVG file. We are
> planning to extend this tool to other types of diagrams such as
> network diagrams, schematic diagrams, and maps with other types of
> modalities. And to achieve this objective we will probably define a
> structured SVG format for these kinds of graphics.
>
> Since you all have been working on similar topics, I am wondering
> whether we can work together?
>
> Cheers,
>
> Chatai

_______________________________________________
dev-accessibility mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-accessibility
Reply | Threaded
Open this post in threaded view
|

Re: Accessible Diagrams

Aaron Leventhal-3
My thoughts on accessible diagrams:

1. We need to introduce new ARIA roles and relations so that objects in
well-known diagram types can describe what they are and their relations
to each other. For example, there could be predefined ARIA semantics for
trees and graphs. I have no SVG skills at the moment, but this is
something like:
<circle role="node" title="Bob McKenzie" ancestorof="employee1 employee2"/>
<square role="node" title="George Schwarzkopf" id="employee1"/>
2. We need role extensibility so that diagram objects can have a type
and describe additional properties and relations, as well as the
necessary localizations.  For some ideas see
http://wiki.mozilla.org/Accessibility/JSON_ARIA
Role extensibility would allow the author to extend past standard
diagram object types but still have a fallback, e.g. <circle
role="manager node"/>
In this case the manager role can define the aria-managerof relation.
3. We need to provide a mechanism to describe units on horizontal and
vertical axes, and allow comparison of objects onto those axes. We need
to be able to see if the diagram objects always snap to discrete points
on the axes or can rome, and whether the horizontal, vertical size or
shape of each object is relevant.
4. We need software to develop accessible diagrams with this markup.
5. We need to introduce a recommended way to process this information
and implement it for Orca or NVDA to show the possibilities. Since the
Mozilla NVDA grant also lists diagram support as a desired outcome it
should not be a problem to work with Mick and Jamie. The Orca team may
be interested in helping as well. However, most likely this is still a
lower priority for both teams and it will require additional external
contributors to get this code in. There needs to be both a sensible
keyboard navigation model as well as some kind of useful review or
summary feedback.

This is all a lot of work, but it's doable if there at least 3 people
serious about making it happen. It would be amazing to finally have the
possibility of accessible diagrams.

- Aaron
_______________________________________________
dev-accessibility mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-accessibility
Reply | Threaded
Open this post in threaded view
|

Re: Accessible Diagrams

David Bolter
In reply to this post by chatai
Hi Cagatay,

As indicated in Aaron's response, there is a timely opportunity here for
you to help inform the specification work as it relates to WAI-ARIA and
SVG. I'd like to participate. How can I find out more about your work?

cheers,
David


chatai wrote:

> Hello Everyone,
>
> My name is Cagatay Goncu, known as Chatai and I am studying PhD at
> Clayton School of Information Technology, Monash University,
> Australia. My research is on accessible diagrams. With this research,
> we are investigating to develop techniques for presenting diagrams in
> different media to blind people in different layouts and modalities
> such as tactile, audio, and tactile/audio.
>
> I am currently working on a tool that gets data from a text input file
> and creates a tactile bar and pie chart as an SVG file. We are
> planning to extend this tool to other types of diagrams such as
> network diagrams, schematic diagrams, and maps with other types of
> modalities. And to achieve this objective we will probably define a
> structured SVG format for these kinds of graphics.
>
> Since you all have been working on similar topics, I am wondering
> whether we can work together?
>
> Cheers,
>
> Chatai
> _______________________________________________
> dev-accessibility mailing list
> [hidden email]
> https://lists.mozilla.org/listinfo/dev-accessibility
>  

_______________________________________________
dev-accessibility mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-accessibility
Reply | Threaded
Open this post in threaded view
|

Re: Accessible Diagrams

Charles McCathieNevile-2
Hi Cagatay,

you should have a look for the work of Philip Roth in this area (if you  
haven't yet). He wrote a PhD 2001 or 2002 that covered pretty much this  
stuff, but rather than diagrams he was looking at illustrative images. He  
did some interesting work on haptic interfaces - but he used a proprietary  
language based on top of windows proprietary software. I spent some time  
with him investigating what SVG (and CSS) needed to make it do the sort of  
things you're talking about, but we never implemented anything on a  
standards-based platform nor published anything very advanced.

Also, his PhD is in French. (It was done at the University of Geneva, and  
if I remember I will be in Norway in two weeks, where I have a copy I can  
look up - ping me about the 27th if you're interested in more info).

cheers

Chaals

(PS Do you know Cam McCormack, who is a major brain behind the Batik SVG  
engine, and is also at Monash? Or meant to be recently departed - maybe he  
finished his thesis now)

On Thu, 17 Apr 2008 19:39:16 +0200, David Bolter  
<[hidden email]> wrote:

> Hi Cagatay,
>
> As indicated in Aaron's response, there is a timely opportunity here for
> you to help inform the specification work as it relates to WAI-ARIA and
> SVG. I'd like to participate. How can I find out more about your work?
>
> cheers,
> David
>
>
> chatai wrote:
>> Hello Everyone,
>>
>> My name is Cagatay Goncu, known as Chatai and I am studying PhD at
>> Clayton School of Information Technology, Monash University,
>> Australia. My research is on accessible diagrams. With this research,
>> we are investigating to develop techniques for presenting diagrams in
>> different media to blind people in different layouts and modalities
>> such as tactile, audio, and tactile/audio.
>>
>> I am currently working on a tool that gets data from a text input file
>> and creates a tactile bar and pie chart as an SVG file. We are
>> planning to extend this tool to other types of diagrams such as
>> network diagrams, schematic diagrams, and maps with other types of
>> modalities. And to achieve this objective we will probably define a
>> structured SVG format for these kinds of graphics.
>>
>> Since you all have been working on similar topics, I am wondering
>> whether we can work together?
>>
>> Cheers,
>>
>> Chatai
>> _______________________________________________
>> dev-accessibility mailing list
>> [hidden email]
>> https://lists.mozilla.org/listinfo/dev-accessibility
>>
>
> _______________________________________________
> dev-accessibility mailing list
> [hidden email]
> https://lists.mozilla.org/listinfo/dev-accessibility



--
Charles McCathieNevile  Opera Software, Standards Group
     je parle français -- hablo español -- jeg lærer norsk
http://my.opera.com/chaals   Try Opera 9.5: http://snapshot.opera.com
_______________________________________________
dev-accessibility mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-accessibility
Reply | Threaded
Open this post in threaded view
|

Re: Accessible Diagrams

Aaron Leventhal-3
In reply to this post by David Bolter
I think there have been at least half a dozen papers written about
making diagrams accessible, but there is still nothing for web authors
to use today. This will become more urgent once SVG becomes easier to
use in HTML. The main thing holding SVG back has been lack of IE support
and lack of integration with text/html. The integration with text/html
is being addressed and thus IE support is likely going to be addressed
eventually as well. In any case, it's possible to make diagrams with CSS
positioned elements.

In this day of open source accessibility and everyone working together
as a community, it would be a shame to see yet another interesting paper
written about diagrams but yet again gain no real benefits for authors
and end users.

That's why I really suggest Catagay look at ARIA as a way of marking
these diagrams up. We can turn this into something real that lasts for a
long time and benefits lots of people.

Catagay, if you want to learn more about ARIA you can look at
http://www.w3.org/WAI/intro/aria.php
We also can answer questions on this newsgroup or in IRC on
irc.w3.org:6665 #wai-aria.

- Aaron



Charles McCathieNevile wrote:

> Hi Cagatay,
>
> you should have a look for the work of Philip Roth in this area (if you
> haven't yet). He wrote a PhD 2001 or 2002 that covered pretty much this
> stuff, but rather than diagrams he was looking at illustrative images.
> He did some interesting work on haptic interfaces - but he used a
> proprietary language based on top of windows proprietary software. I
> spent some time with him investigating what SVG (and CSS) needed to make
> it do the sort of things you're talking about, but we never implemented
> anything on a standards-based platform nor published anything very
> advanced.
>
> Also, his PhD is in French. (It was done at the University of Geneva,
> and if I remember I will be in Norway in two weeks, where I have a copy
> I can look up - ping me about the 27th if you're interested in more info).
>
> cheers
>
> Chaals
>
> (PS Do you know Cam McCormack, who is a major brain behind the Batik SVG
> engine, and is also at Monash? Or meant to be recently departed - maybe
> he finished his thesis now)
>
> On Thu, 17 Apr 2008 19:39:16 +0200, David Bolter
> <[hidden email]> wrote:
>
>> Hi Cagatay,
>>
>> As indicated in Aaron's response, there is a timely opportunity here for
>> you to help inform the specification work as it relates to WAI-ARIA and
>> SVG. I'd like to participate. How can I find out more about your work?
>>
>> cheers,
>> David
>>
>>
>> chatai wrote:
>>> Hello Everyone,
>>>
>>> My name is Cagatay Goncu, known as Chatai and I am studying PhD at
>>> Clayton School of Information Technology, Monash University,
>>> Australia. My research is on accessible diagrams. With this research,
>>> we are investigating to develop techniques for presenting diagrams in
>>> different media to blind people in different layouts and modalities
>>> such as tactile, audio, and tactile/audio.
>>>
>>> I am currently working on a tool that gets data from a text input file
>>> and creates a tactile bar and pie chart as an SVG file. We are
>>> planning to extend this tool to other types of diagrams such as
>>> network diagrams, schematic diagrams, and maps with other types of
>>> modalities. And to achieve this objective we will probably define a
>>> structured SVG format for these kinds of graphics.
>>>
>>> Since you all have been working on similar topics, I am wondering
>>> whether we can work together?
>>>
>>> Cheers,
>>>
>>> Chatai
>>> _______________________________________________
>>> dev-accessibility mailing list
>>> [hidden email]
>>> https://lists.mozilla.org/listinfo/dev-accessibility
>>>
>>
>> _______________________________________________
>> dev-accessibility mailing list
>> [hidden email]
>> https://lists.mozilla.org/listinfo/dev-accessibility
>
>
>


_______________________________________________
dev-accessibility mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-accessibility
Reply | Threaded
Open this post in threaded view
|

Re: Accessible Diagrams

David Bolter
Hi Aaron, I agree on all points.

Catagay, is this a goal of your research?  To produce real, near-future
results?  Or is it distant-future?

If it is near future, I'd like to chat about collaboration :)

cheers,
David

Aaron Leventhal wrote:

> I think there have been at least half a dozen papers written about
> making diagrams accessible, but there is still nothing for web authors
> to use today. This will become more urgent once SVG becomes easier to
> use in HTML. The main thing holding SVG back has been lack of IE
> support and lack of integration with text/html. The integration with
> text/html is being addressed and thus IE support is likely going to be
> addressed eventually as well. In any case, it's possible to make
> diagrams with CSS positioned elements.
>
> In this day of open source accessibility and everyone working together
> as a community, it would be a shame to see yet another interesting
> paper written about diagrams but yet again gain no real benefits for
> authors and end users.
>
> That's why I really suggest Catagay look at ARIA as a way of marking
> these diagrams up. We can turn this into something real that lasts for
> a long time and benefits lots of people.
>
> Catagay, if you want to learn more about ARIA you can look at
> http://www.w3.org/WAI/intro/aria.php
> We also can answer questions on this newsgroup or in IRC on
> irc.w3.org:6665 #wai-aria.
>
> - Aaron
>
>
>
> Charles McCathieNevile wrote:
>> Hi Cagatay,
>>
>> you should have a look for the work of Philip Roth in this area (if you
>> haven't yet). He wrote a PhD 2001 or 2002 that covered pretty much this
>> stuff, but rather than diagrams he was looking at illustrative images.
>> He did some interesting work on haptic interfaces - but he used a
>> proprietary language based on top of windows proprietary software. I
>> spent some time with him investigating what SVG (and CSS) needed to make
>> it do the sort of things you're talking about, but we never implemented
>> anything on a standards-based platform nor published anything very
>> advanced.
>>
>> Also, his PhD is in French. (It was done at the University of Geneva,
>> and if I remember I will be in Norway in two weeks, where I have a copy
>> I can look up - ping me about the 27th if you're interested in more
>> info).
>>
>> cheers
>>
>> Chaals
>>
>> (PS Do you know Cam McCormack, who is a major brain behind the Batik SVG
>> engine, and is also at Monash? Or meant to be recently departed - maybe
>> he finished his thesis now)
>>
>> On Thu, 17 Apr 2008 19:39:16 +0200, David Bolter
>> <[hidden email]> wrote:
>>
>>> Hi Cagatay,
>>>
>>> As indicated in Aaron's response, there is a timely opportunity here
>>> for
>>> you to help inform the specification work as it relates to WAI-ARIA and
>>> SVG. I'd like to participate. How can I find out more about your work?
>>>
>>> cheers,
>>> David
>>>
>>>
>>> chatai wrote:
>>>> Hello Everyone,
>>>>
>>>> My name is Cagatay Goncu, known as Chatai and I am studying PhD at
>>>> Clayton School of Information Technology, Monash University,
>>>> Australia. My research is on accessible diagrams. With this research,
>>>> we are investigating to develop techniques for presenting diagrams in
>>>> different media to blind people in different layouts and modalities
>>>> such as tactile, audio, and tactile/audio.
>>>>
>>>> I am currently working on a tool that gets data from a text input file
>>>> and creates a tactile bar and pie chart as an SVG file. We are
>>>> planning to extend this tool to other types of diagrams such as
>>>> network diagrams, schematic diagrams, and maps with other types of
>>>> modalities. And to achieve this objective we will probably define a
>>>> structured SVG format for these kinds of graphics.
>>>>
>>>> Since you all have been working on similar topics, I am wondering
>>>> whether we can work together?
>>>>
>>>> Cheers,
>>>>
>>>> Chatai
>>>> _______________________________________________
>>>> dev-accessibility mailing list
>>>> [hidden email]
>>>> https://lists.mozilla.org/listinfo/dev-accessibility
>>>>
>>>
>>> _______________________________________________
>>> dev-accessibility mailing list
>>> [hidden email]
>>> https://lists.mozilla.org/listinfo/dev-accessibility
>>
>>
>>
>
>

_______________________________________________
dev-accessibility mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-accessibility
Reply | Threaded
Open this post in threaded view
|

Re: Accessible Diagrams

chatai
In reply to this post by Aaron Leventhal-3
Hello Everyone,

First of all thank you very much for all your comments.

I think I need to give more information about what I am doing and
planning to do. I have started my PhD about 7 months ago. So, I can
say that I am pretty new in this area, and it looks like that I need
to read in more detail about ARIA, AT API...etc. I have heard all
these before but I couldnt pass to the detailed sections of them. I
will probably ask lots of questions to you :)

As I mentioned before I am currently working on a tactile chart tool.
We are collaborating with Vision Australia in the development of this
tool. Actually, for now they are the first people that will use the
tool. Tomorrow we are going to give a demo of the tool, and get some
feedback from them... (After the demo I will put the tool on the
internet, I hope it works as a Java applet)

The next step will be doing some user studies with blind people to
understand their preferences about different layouts of bar charts. In
these user studies we are planning to adapt the charts slightly to
find out the important aspects of adaptations.

Our final aim is to find a way to include the semantic information of
a graphic into itself. It looks like that the best way to do it is
using SVG and defining the semnatics within the SVG file. I think that
ARIA can be a good choice for this purpose. Also I think that the
generation of the roles and relations can be (semi)automated for each
type of diagram. For instance, in an organization chart putting a
shape under another shape can give the ordinate/subordinate relation
and the role of each item in the relation. Thus, the spatial relations
among the items in graphics can give lots of semantic information.

My supervisor has just brought me some updates regarding to the demo :
(  So, I need to do them, I will write you in more detail
tomorrrow... ;)

Cheers,

Chatai

On Apr 18, 11:23 pm, David Bolter <[hidden email]> wrote:

> Hi Aaron, I agree on all points.
>
> Catagay, is this a goal of your research?  To produce real, near-future
> results?  Or is it distant-future?
>
> If it is near future, I'd like to chat about collaboration :)
>
> cheers,
> David
>
> Aaron Leventhal wrote:
> > I think there have been at least half a dozen papers written about
> > making diagrams accessible, but there is still nothing for web authors
> > to use today. This will become more urgent once SVG becomes easier to
> > use in HTML. The main thing holding SVG back has been lack of IE
> > support and lack of integration with text/html. The integration with
> > text/html is being addressed and thus IE support is likely going to be
> > addressed eventually as well. In any case, it's possible to make
> > diagrams with CSS positioned elements.
>
> > In this day of open source accessibility and everyone working together
> > as a community, it would be a shame to see yet another interesting
> > paper written about diagrams but yet again gain no real benefits for
> > authors and end users.
>
> > That's why I really suggest Catagay look at ARIA as a way of marking
> > these diagrams up. We can turn this into something real that lasts for
> > a long time and benefits lots of people.
>
> > Catagay, if you want to learn more about ARIA you can look at
> >http://www.w3.org/WAI/intro/aria.php
> > We also can answer questions on this newsgroup or in IRC on
> > irc.w3.org:6665 #wai-aria.
>
> > - Aaron
>
> > Charles McCathieNevile wrote:
> >> Hi Cagatay,
>
> >> you should have a look for the work of Philip Roth in this area (if you
> >> haven't yet). He wrote a PhD 2001 or 2002 that covered pretty much this
> >> stuff, but rather than diagrams he was looking at illustrative images.
> >> He did some interesting work on haptic interfaces - but he used a
> >> proprietary language based on top of windows proprietary software. I
> >> spent some time with him investigating what SVG (and CSS) needed to make
> >> it do the sort of things you're talking about, but we never implemented
> >> anything on a standards-based platform nor published anything very
> >> advanced.
>
> >> Also, his PhD is in French. (It was done at the University of Geneva,
> >> and if I remember I will be in Norway in two weeks, where I have a copy
> >> I can look up - ping me about the 27th if you're interested in more
> >> info).
>
> >> cheers
>
> >> Chaals
>
> >> (PS Do you know Cam McCormack, who is a major brain behind the Batik SVG
> >> engine, and is also at Monash? Or meant to be recently departed - maybe
> >> he finished his thesis now)
>
> >> On Thu, 17 Apr 2008 19:39:16 +0200, David Bolter
> >> <[hidden email]> wrote:
>
> >>> Hi Cagatay,
>
> >>> As indicated in Aaron's response, there is a timely opportunity here
> >>> for
> >>> you to help inform the specification work as it relates to WAI-ARIA and
> >>> SVG. I'd like to participate. How can I find out more about your work?
>
> >>> cheers,
> >>> David
>
> >>> chatai wrote:
> >>>> Hello Everyone,
>
> >>>> My name is Cagatay Goncu, known as Chatai and I am studying PhD at
> >>>> Clayton School of Information Technology, Monash University,
> >>>> Australia. My research is on accessible diagrams. With this research,
> >>>> we are investigating to develop techniques for presenting diagrams in
> >>>> different media to blind people in different layouts and modalities
> >>>> such as tactile, audio, and tactile/audio.
>
> >>>> I am currently working on a tool that gets data from a text input file
> >>>> and creates a tactile bar and pie chart as an SVG file. We are
> >>>> planning to extend this tool to other types of diagrams such as
> >>>> network diagrams, schematic diagrams, and maps with other types of
> >>>> modalities. And to achieve this objective we will probably define a
> >>>> structured SVG format for these kinds of graphics.
>
> >>>> Since you all have been working on similar topics, I am wondering
> >>>> whether we can work together?
>
> >>>> Cheers,
>
> >>>> Chatai
> >>>> _______________________________________________
> >>>> dev-accessibility mailing list
> >>>> [hidden email]
> >>>>https://lists.mozilla.org/listinfo/dev-accessibility
>
> >>> _______________________________________________
> >>> dev-accessibility mailing list
> >>> [hidden email]
> >>>https://lists.mozilla.org/listinfo/dev-accessibility

_______________________________________________
dev-accessibility mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-accessibility
Reply | Threaded
Open this post in threaded view
|

Re: Accessible Diagrams

Aaron Leventhal-3
Cagatay,

You are looking at hard copy tactile output, but not text to speech via
a screen reader?

- Aaron
_______________________________________________
dev-accessibility mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-accessibility
Reply | Threaded
Open this post in threaded view
|

Re: Accessible Diagrams

Al Gilman
In reply to this post by chatai
Hi, Chatai.

I want to offer you some cautions as regards the available Web
technolgies, and some further reading opportunities <groan/>.

[Yes, please work with W3C/WAI on roles for diagram
content, but...]

SVG has much better support for metadata in its basic structure
than does HTML.  So a lot of what we have done in WAI-ARIA to dance
around the limitations of HTML doesn't have to be necessary if
your SVG is used as an intermediate language between the content
analyzer and the synthesis phase that generates the haptic-graphics
[interactive] scene.  You may also want to use SVG for a device-aware
presentation format with some of the user-side-adjustment preserved.

 From a math and science point of view,  in Web 2.0 interaction
we have a tree-based content world with a light spiderweb overlay
of relationship markup that makes the full graph more connected
than just a tree.  But the syntax tree dominates.

In diagrams, the graph  nature is more central, and some of the
compromises we have made in ARIA may not fit so well.  In particular,
we use generic/specific relationships in the role ontology that
is used to qualify the processing of *elements* but not a similar
ontology for *attributes, including relationships*.  For the attributes,
we just have standard-defined unary properties and binary relationships.

For things like the ordinate/subordinate relationship overlay, we
may find that this should be embedded in a relationship ontology, that
is to say use the "sub-relation" relationship as captured in OWL.

There have been a variety of devices developed over the years to make
haptic exploration of scenes more interactive, from a touch-screen
workalike that takes an embossed diagram overlay to a mouse with
vibrating dots that somewhat gives an Optacon scan of a virtual image
or diagram available as data in a computer.

So you don't want your scene analysis to presume that all you have
between you and the user is a static embossed sheet.  Access to more
than you can represent statically in the haptic graphics may be afforded
through event-driven system responses where the user events are cued
by what the user feels in the haptic imagery.  Even the point and click
world is not finding that they what to, in a transient way, enrich the
scene in the logical vicinity of where the user seems to be thinking
as reflected by their mouse point.

In haptics we have fewer resolution cells; we should anticipate the
interaction to be less asymmetrical, with a larger share of user
initiative in proportion to system initiative (even if haptics is
graphic in a way that audio out is hardly... but I digress).

The work on virtual keyboards that use gesture recognition would also
suggest that it is easy to create the touch-screen effect with a
haptic diagram and a touchless touchscreen device watching for taps
on the diagram.

On to the reading list:

a) the record of the "access to visualization" telecon that the WAI
put on:

http://www.w3.org/WAI/RD/2004/06/28-agenda.html

In particular, I would emphasize the idea that what you want do do
in your analysis phase is deconstruct the diagram, or get in bed with
the application that is behind the diagram, to capture what is expressed
in the diagram in "the information model of the application world."

This is your universal-design strategy for then targeting an
[interactive] presentation to the interaction world that the user's
devices and skills support well.

in addition to all the references I raised there...

b) don't be ignorant of what Will Pearson is doing.  He is further
along the Ph.D. track and working in a related area.  Don't get
caught up short re-inventing what he has just done.

[hidden email]

c) consider the SUPPLE optimizing layout adaptor for the generation
of your haptic imagery.  You have a more constrained problem than the
designer of the graphics targeted to the eyeball.  You have the same
amount of information in the application sense and far fewer "minimum
feature size"-es in your "effective screen width."  So an optimum
packing of the discernible features, guided by application-driven
topology but not in the end constrained by visible-graphics-driven
geometry -- could be of benefit in this application.

http://dada.cs.washington.edu/supple/index.php/Main_Page

Oh, yes; and

d) our whitepaper on multimodal interaction and adaptation:

http://lists.w3.org/Archives/Public/www-multimodal/2007Jun/att-0002/ 
Accessibility_Notes_on_MMI_Architecture.html

Al

On 21 Apr 2008, at 2:18 AM, Cagatay Goncu (Chatai) wrote:

> Hello Everyone,
>
> First of all thank you very much for all your comments.
>
> I think I need to give more information about what I am doing and
> planning to do. I have started my PhD about 7 months ago. So, I can
> say that I am pretty new in this area, and it looks like that I need
> to read in more detail about ARIA, AT API...etc. I have heard all
> these before but I couldnt pass to the detailed sections of them. I
> will probably ask lots of questions to you :)
>
> As I mentioned before I am currently working on a tactile chart tool.
> We are collaborating with Vision Australia in the development of this
> tool. Actually, for now they are the first people that will use the
> tool. Tomorrow we are going to give a demo of the tool, and get some
> feedback from them... (After the demo I will put the tool on the
> internet, I hope it works as a Java applet)
>
> The next step will be doing some user studies with blind people to
> understand their preferences about different layouts of bar charts. In
> these user studies we are planning to adapt the charts slightly to
> find out the important aspects of adaptations.
>
> Our final aim is to find a way to include the semantic information of
> a graphic into itself. It looks like that the best way to do it is
> using SVG and defining the semnatics within the SVG file. I think that
> ARIA can be a good choice for this purpose. Also I think that the
> generation of the roles and relations can be (semi)automated for each
> type of diagram. For instance, in an organization chart putting a
> shape under another shape can give the ordinate/subordinate relation
> and the role of each item in the relation. Thus, the spatial relations
> among the items in graphics can give lots of semantic information.
>
> My supervisor has just brought me some updates regarding to the demo :
> (  So, I need to do them, I will write you in more detail
> tomorrrow... ;)
>
> Cheers,
>
> Chatai
>
> On Apr 18, 11:23 pm, David Bolter <[hidden email]> wrote:
>> Hi Aaron, I agree on all points.
>>
>> Catagay, is this a goal of your research?  To produce real, near-
>> future
>> results?  Or is it distant-future?
>>
>> If it is near future, I'd like to chat about collaboration :)
>>
>> cheers,
>> David
>>
>> Aaron Leventhal wrote:
>>> I think there have been at least half a dozen papers written about
>>> making diagrams accessible, but there is still nothing for web  
>>> authors
>>> to use today. This will become more urgent once SVG becomes  
>>> easier to
>>> use in HTML. The main thing holding SVG back has been lack of IE
>>> support and lack of integration with text/html. The integration with
>>> text/html is being addressed and thus IE support is likely going  
>>> to be
>>> addressed eventually as well. In any case, it's possible to make
>>> diagrams with CSS positioned elements.
>>
>>> In this day of open source accessibility and everyone working  
>>> together
>>> as a community, it would be a shame to see yet another interesting
>>> paper written about diagrams but yet again gain no real benefits for
>>> authors and end users.
>>
>>> That's why I really suggest Catagay look at ARIA as a way of marking
>>> these diagrams up. We can turn this into something real that  
>>> lasts for
>>> a long time and benefits lots of people.
>>
>>> Catagay, if you want to learn more about ARIA you can look at
>>> http://www.w3.org/WAI/intro/aria.php
>>> We also can answer questions on this newsgroup or in IRC on
>>> irc.w3.org:6665 #wai-aria.
>>
>>> - Aaron
>>
>>> Charles McCathieNevile wrote:
>>>> Hi Cagatay,
>>
>>>> you should have a look for the work of Philip Roth in this area  
>>>> (if you
>>>> haven't yet). He wrote a PhD 2001 or 2002 that covered pretty  
>>>> much this
>>>> stuff, but rather than diagrams he was looking at illustrative  
>>>> images.
>>>> He did some interesting work on haptic interfaces - but he used a
>>>> proprietary language based on top of windows proprietary  
>>>> software. I
>>>> spent some time with him investigating what SVG (and CSS) needed  
>>>> to make
>>>> it do the sort of things you're talking about, but we never  
>>>> implemented
>>>> anything on a standards-based platform nor published anything very
>>>> advanced.
>>
>>>> Also, his PhD is in French. (It was done at the University of  
>>>> Geneva,
>>>> and if I remember I will be in Norway in two weeks, where I have  
>>>> a copy
>>>> I can look up - ping me about the 27th if you're interested in more
>>>> info).
>>
>>>> cheers
>>
>>>> Chaals
>>
>>>> (PS Do you know Cam McCormack, who is a major brain behind the  
>>>> Batik SVG
>>>> engine, and is also at Monash? Or meant to be recently departed  
>>>> - maybe
>>>> he finished his thesis now)
>>
>>>> On Thu, 17 Apr 2008 19:39:16 +0200, David Bolter
>>>> <[hidden email]> wrote:
>>
>>>>> Hi Cagatay,
>>
>>>>> As indicated in Aaron's response, there is a timely opportunity  
>>>>> here
>>>>> for
>>>>> you to help inform the specification work as it relates to WAI-
>>>>> ARIA and
>>>>> SVG. I'd like to participate. How can I find out more about  
>>>>> your work?
>>
>>>>> cheers,
>>>>> David
>>
>>>>> chatai wrote:
>>>>>> Hello Everyone,
>>
>>>>>> My name is Cagatay Goncu, known as Chatai and I am studying  
>>>>>> PhD at
>>>>>> Clayton School of Information Technology, Monash University,
>>>>>> Australia. My research is on accessible diagrams. With this  
>>>>>> research,
>>>>>> we are investigating to develop techniques for presenting  
>>>>>> diagrams in
>>>>>> different media to blind people in different layouts and  
>>>>>> modalities
>>>>>> such as tactile, audio, and tactile/audio.
>>
>>>>>> I am currently working on a tool that gets data from a text  
>>>>>> input file
>>>>>> and creates a tactile bar and pie chart as an SVG file. We are
>>>>>> planning to extend this tool to other types of diagrams such as
>>>>>> network diagrams, schematic diagrams, and maps with other  
>>>>>> types of
>>>>>> modalities. And to achieve this objective we will probably  
>>>>>> define a
>>>>>> structured SVG format for these kinds of graphics.
>>
>>>>>> Since you all have been working on similar topics, I am wondering
>>>>>> whether we can work together?
>>
>>>>>> Cheers,
>>
>>>>>> Chatai
>>>>>> _______________________________________________
>>>>>> dev-accessibility mailing list
>>>>>> [hidden email]
>>>>>> https://lists.mozilla.org/listinfo/dev-accessibility
>>
>>>>> _______________________________________________
>>>>> dev-accessibility mailing list
>>>>> [hidden email]
>>>>> https://lists.mozilla.org/listinfo/dev-accessibility
>
> _______________________________________________
> dev-accessibility mailing list
> [hidden email]
> https://lists.mozilla.org/listinfo/dev-accessibility

_______________________________________________
dev-accessibility mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-accessibility
Reply | Threaded
Open this post in threaded view
|

Re: Accessible Diagrams

Steve Lee-3
On 21/04/2008, Al Gilman <[hidden email]> wrote:
>  The work on virtual keyboards that use gesture recognition would also
>  suggest that it is easy to create the touch-screen effect with a
>  haptic diagram and a touchless touchscreen device watching for taps
>  on the diagram.

Al, can you provide references/URLs to that work? I'm interested from
an alternative input angle.

Thanks

--
Steve Lee
--
Open Source Assistive Technology Software
web: fullmeasure.co.uk
blog: eduspaces.net/stevelee/weblog
_______________________________________________
dev-accessibility mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-accessibility
Reply | Threaded
Open this post in threaded view
|

Re: Accessible Diagrams

Al Gilman

On 21 Apr 2008, at 11:21 AM, Steve Lee wrote:

> On 21/04/2008, Al Gilman <[hidden email]> wrote:
>>  The work on virtual keyboards that use gesture recognition would  
>> also
>>  suggest that it is easy to create the touch-screen effect with a
>>  haptic diagram and a touchless touchscreen device watching for taps
>>  on the diagram.
>
> Al, can you provide references/URLs to that work? I'm interested from
> an alternative input angle.

In no particular order, and definitely not exhaustive, here's what I  
could reconstruct with a few quick searches.

http://w4.siemens.de/FuI/en/archiv/zeitschrift/heft1_98/artikel07/

http://www.senseboard.com/index.php

my best search term set to date is

http://www.google.com/search?q=gesture+recognition+keyboard

Gregg had a commercial device for people to play with at the
2007 State of the Science meeting, but there is no report from
that meeting on the website that I found.

Al

>
> Thanks
>
> --
> Steve Lee
> --
> Open Source Assistive Technology Software
> web: fullmeasure.co.uk
> blog: eduspaces.net/stevelee/weblog
> _______________________________________________
> dev-accessibility mailing list
> [hidden email]
> https://lists.mozilla.org/listinfo/dev-accessibility

_______________________________________________
dev-accessibility mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-accessibility
Reply | Threaded
Open this post in threaded view
|

Re: Accessible Diagrams

chatai
In reply to this post by Aaron Leventhal-3
On Apr 21, 10:21 pm, Aaron Leventhal <[hidden email]>
wrote:
> Cagatay,
>
> You are looking at hard copy tactile output, but not text to speech via
> a screen reader?
>
> - Aaron

Yes we are planning to make user studies with hard copy tactile output
next month...

But I think using audio and speech is a better way. So, after this
user study we will probably focus on audio and speech... I think
tactile interaction is also important, and it can be used in
navigation within the audio and speech output, like using tactile
gestures...

Whether it is audio/speech or tactile , I think finding a way to
include semantics within graphics is the most important point. Because
the graphics should be adapted according to the needs of blind users.
And these adaptations need both semantics and structure of the
graphics... That's why I think that ARIA can be a good way to do
this...

chatai
_______________________________________________
dev-accessibility mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-accessibility
Reply | Threaded
Open this post in threaded view
|

Re: Accessible Diagrams

chatai
In reply to this post by chatai
On Apr 22, 12:53 am, Al Gilman <[hidden email]> wrote:

> Hi, Chatai.
>
> I want to offer you some cautions as regards the available Web
> technolgies, and some further reading opportunities <groan/>.
>
> [Yes, please work with W3C/WAI on roles for diagram
> content, but...]
>
> SVG has much better support for metadata in its basic structure
> than does HTML.  So a lot of what we have done in WAI-ARIA to dance
> around the limitations of HTML doesn't have to be necessary if
> your SVG is used as an intermediate language between the content
> analyzer and the synthesis phase that generates the haptic-graphics
> [interactive] scene.  You may also want to use SVG for a device-aware
> presentation format with some of the user-side-adjustment preserved.
>
>  From a math and science point of view,  in Web 2.0 interaction
> we have a tree-based content world with a light spiderweb overlay
> of relationship markup that makes the full graph more connected
> than just a tree.  But the syntax tree dominates.
>
> In diagrams, the graph  nature is more central, and some of the
> compromises we have made in ARIA may not fit so well.  In particular,
> we use generic/specific relationships in the role ontology that
> is used to qualify the processing of *elements* but not a similar
> ontology for *attributes, including relationships*.  For the attributes,
> we just have standard-defined unary properties and binary relationships.
>
> For things like the ordinate/subordinate relationship overlay, we
> may find that this should be embedded in a relationship ontology, that
> is to say use the "sub-relation" relationship as captured in OWL.
>
> There have been a variety of devices developed over the years to make
> haptic exploration of scenes more interactive, from a touch-screen
> workalike that takes an embossed diagram overlay to a mouse with
> vibrating dots that somewhat gives an Optacon scan of a virtual image
> or diagram available as data in a computer.
>
> So you don't want your scene analysis to presume that all you have
> between you and the user is a static embossed sheet.  Access to more
> than you can represent statically in the haptic graphics may be afforded
> through event-driven system responses where the user events are cued
> by what the user feels in the haptic imagery.  Even the point and click
> world is not finding that they what to, in a transient way, enrich the
> scene in the logical vicinity of where the user seems to be thinking
> as reflected by their mouse point.
>
> In haptics we have fewer resolution cells; we should anticipate the
> interaction to be less asymmetrical, with a larger share of user
> initiative in proportion to system initiative (even if haptics is
> graphic in a way that audio out is hardly... but I digress).
>
> The work on virtual keyboards that use gesture recognition would also
> suggest that it is easy to create the touch-screen effect with a
> haptic diagram and a touchless touchscreen device watching for taps
> on the diagram.
>
> On to the reading list:
>
> a) the record of the "access to visualization" telecon that the WAI
> put on:
>
> http://www.w3.org/WAI/RD/2004/06/28-agenda.html
>
> In particular, I would emphasize the idea that what you want do do
> in your analysis phase is deconstruct the diagram, or get in bed with
> the application that is behind the diagram, to capture what is expressed
> in the diagram in "the information model of the application world."
>
> This is your universal-design strategy for then targeting an
> [interactive] presentation to the interaction world that the user's
> devices and skills support well.
>
> in addition to all the references I raised there...
>
> b) don't be ignorant of what Will Pearson is doing.  He is further
> along the Ph.D. track and working in a related area.  Don't get
> caught up short re-inventing what he has just done.
>
> [hidden email]
>
> c) consider the SUPPLE optimizing layout adaptor for the generation
> of your haptic imagery.  You have a more constrained problem than the
> designer of the graphics targeted to the eyeball.  You have the same
> amount of information in the application sense and far fewer "minimum
> feature size"-es in your "effective screen width."  So an optimum
> packing of the discernible features, guided by application-driven
> topology but not in the end constrained by visible-graphics-driven
> geometry -- could be of benefit in this application.
>
> http://dada.cs.washington.edu/supple/index.php/Main_Page
>
> Oh, yes; and
>
> d) our whitepaper on multimodal interaction and adaptation:
>
> http://lists.w3.org/Archives/Public/www-multimodal/2007Jun/att-0002/
> Accessibility_Notes_on_MMI_Architecture.html
>
> Al
>
> On 21 Apr 2008, at 2:18 AM, Cagatay Goncu (Chatai) wrote:
>
> > Hello Everyone,
>
> > First of all thank you very much for all your comments.
>
> > I think I need to give more information about what I am doing and
> > planning to do. I have started my PhD about 7 months ago. So, I can
> > say that I am pretty new in this area, and it looks like that I need
> > to read in more detail about ARIA, AT API...etc. I have heard all
> > these before but I couldnt pass to the detailed sections of them. I
> > will probably ask lots of questions to you :)
>
> > As I mentioned before I am currently working on a tactile chart tool.
> > We are collaborating with Vision Australia in the development of this
> > tool. Actually, for now they are the first people that will use the
> > tool. Tomorrow we are going to give a demo of the tool, and get some
> > feedback from them... (After the demo I will put the tool on the
> > internet, I hope it works as a Java applet)
>
> > The next step will be doing some user studies with blind people to
> > understand their preferences about different layouts of bar charts. In
> > these user studies we are planning to adapt the charts slightly to
> > find out the important aspects of adaptations.
>
> > Our final aim is to find a way to include the semantic information of
> > a graphic into itself. It looks like that the best way to do it is
> > using SVG and defining the semnatics within the SVG file. I think that
> > ARIA can be a good choice for this purpose. Also I think that the
> > generation of the roles and relations can be (semi)automated for each
> > type of diagram. For instance, in an organization chart putting a
> > shape under another shape can give the ordinate/subordinate relation
> > and the role of each item in the relation. Thus, the spatial relations
> > among the items in graphics can give lots of semantic information.
>
> > My supervisor has just brought me some updates regarding to the demo :
> > (  So, I need to do them, I will write you in more detail
> > tomorrrow... ;)
>
> > Cheers,
>
> > Chatai
>
> > On Apr 18, 11:23 pm, David Bolter <[hidden email]> wrote:
> >> Hi Aaron, I agree on all points.
>
> >> Catagay, is this a goal of your research?  To produce real, near-
> >> future
> >> results?  Or is it distant-future?
>
> >> If it is near future, I'd like to chat about collaboration :)
>
> >> cheers,
> >> David
>
> >> Aaron Leventhal wrote:
> >>> I think there have been at least half a dozen papers written about
> >>> making diagrams accessible, but there is still nothing for web
> >>> authors
> >>> to use today. This will become more urgent once SVG becomes
> >>> easier to
> >>> use in HTML. The main thing holding SVG back has been lack of IE
> >>> support and lack of integration with text/html. The integration with
> >>> text/html is being addressed and thus IE support is likely going
> >>> to be
> >>> addressed eventually as well. In any case, it's possible to make
> >>> diagrams with CSS positioned elements.
>
> >>> In this day of open source accessibility and everyone working
> >>> together
> >>> as a community, it would be a shame to see yet another interesting
> >>> paper written about diagrams but yet again gain no real benefits for
> >>> authors and end users.
>
> >>> That's why I really suggest Catagay look at ARIA as a way of marking
> >>> these diagrams up. We can turn this into something real that
> >>> lasts for
> >>> a long time and benefits lots of people.
>
> >>> Catagay, if you want to learn more about ARIA you can look at
> >>>http://www.w3.org/WAI/intro/aria.php
> >>> We also can answer questions on this newsgroup or in IRC on
> >>> irc.w3.org:6665 #wai-aria.
>
> >>> - Aaron
>
> >>> Charles McCathieNevile wrote:
> >>>> Hi Cagatay,
>
> >>>> you should have a look for the work of Philip Roth in this area
> >>>> (if you
> >>>> haven't yet). He wrote a PhD 2001 or 2002 that covered pretty
> >>>> much this
> >>>> stuff, but rather than diagrams he was looking at illustrative
> >>>> images.
> >>>> He did some interesting work on haptic interfaces - but he used a
> >>>> proprietary language based on top of windows proprietary
> >>>> software. I
> >>>> spent some time with him investigating what SVG (and CSS) needed
> >>>> to make
> >>>> it do the sort of things you're talking about, but we never
> >>>> implemented
> >>>> anything on a standards-based platform nor published anything very
> >>>> advanced.
>
> >>>> Also, his PhD is in French. (It was done at the University of
> >>>> Geneva,
> >>>> and if I remember I will be in Norway in two weeks, where I have
> >>>> a copy
> >>>> I can look up - ping me about the 27th if you're interested in more
> >>>> info).
>
> >>>> cheers
>
> >>>> Chaals
>
> >>>> (PS Do you know Cam McCormack, who is a major brain behind the
> >>>> Batik SVG
> >>>> engine, and is also at Monash? Or meant to be recently departed
> >>>> - maybe
> >>>> he finished his thesis now)
>
> >>>> On Thu, 17 Apr 2008 19:39:16 +0200, David Bolter
> >>>> <[hidden email]> wrote:
>
> >>>>> Hi Cagatay,
>
> >>>>> As indicated in Aaron's response, there is a timely opportunity
> >>>>> here
> >>>>> for
> >>>>> you to help inform the specification work as it relates to WAI-
> >>>>> ARIA and
> >>>>> SVG. I'd like to participate. How can I find out more about
> >>>>> your work?
>
> >>>>> cheers,
> >>>>> David
>
> >>>>> chatai wrote:
> >>>>>> Hello Everyone,
>
> >>>>>> My name is Cagatay Goncu, known as Chatai and I am studying
> >>>>>> PhD at
> >>>>>> Clayton School of Information Technology, Monash University,
> >>>>>> Australia. My research is on accessible diagrams. With this
>
> ...
>
> read more »

Thank you Al for your feedback and the links...

Now I have more to read. ..

Cheers,

Chatai
_______________________________________________
dev-accessibility mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-accessibility