W3C home > Mailing lists > Public > w3c-wai-ua@w3.org > July to September 2001

Re: Is this a Job for SVG and WAI?

From: Al Gilman <asgilman@iamdigex.net>
Date: Sat, 18 Aug 2001 06:06:49 -0400
Message-Id: <200108180154.VAA5902705@smtp2.mail.iamworld.net>
To: Jon Ferraiolo <jferraio@adobe.com>, Jon Gunderson <jongund@uiuc.edu>
Cc: <w3c-svg-wg@w3.org>, <w3c-wai-ua@w3.org>
Two updates:

a) Charles is not so available as I had hoped.

b) I looked at Diggles's poster.

Here is a rough outline of the state of the art in this matter:

The poster as a whole:

Is not something you would want to move to the Web as a poster layout.  Or not
as the poster layout alone.  This is a pretty awkward arrangement of the
contents if you have an interactive workstation to work with you, even for the
sighted.  The accessible transcription of this poster would generally follow
the structure guidelines of a a digital talking book. 

 NISO Working Papers
 <http://www.loc.gov/nls/niso>http://www.loc.gov/nls/niso/
 
You could have a poster XSLT that formatted the whole thing on one wall
chart. 
But the content would be in a more "pool of resources with multiple possible
tours and layouts" format.  Device independent, if you will.  What to do with
the material is reasonably clear from the talking book specification, outside
of the four figures.

Figures #1 and #2 can go into accessible SVG by the techniques already
discussed in the existing "Accessibility features of SVG."  This conversion is
a SMOP, not a question of devising new strategies.  

Figures #3 and #4 contain sensor imagery and how to handle the image
content is
a research topic.  The highest and best access is that the online version is a
data browser that has the capabiblities to generate all these figures from the
multisensor data bank and the figures are just presets cached in the data
browser, but the user can generate any other view in terms both of information
selection and representation within the capabilities of the data browser which
include binding to haptics, SVG, what have you.  In a system like this, the
heavy lifting is done prior to reduction to SVG in systems such as the
scalable
visualization toolkit

 <http://vistools.npaci.edu/home.html>http://vistools.npaci.edu/home.html

But SVG could be a good choice as the Web intermediate language for client
communication even in an interactive data browse scenario such as this.  

There are a variety of accessibility applications for the resulting software
architecture.  Adding morphologically-registered highlights to real-time
speaker video so as to provide cued speech for speech readers, for example.

 The role of User Environments in The Grid
 <http://trace.wisc.edu/docs/ud4grid/#_Toc495220373>http://trace.wisc.edu/d
ocs/ud4grid/#_Toc495220373

For use of an SVG-breathing remote web device as the client workstation in
such
a system, we could use SVG as an intermediate language to share with the Web
device; but this involves a 2-1/2 dimensional model (multiple values of
z-index, multiple layers of 2D stuff) I believe to pull off.  For example the
visual and the haptic data would make sense to put in independent SVG
layers or
independent layers of which the visual was SVG but the registration was rock
solid by the fact that both are built on registered box frames.  It is
possible
this can be done with just one graphic layer and rich metadata encoded in RDF
and included via the svg:metadata element.

The strategies that one can use today are pretty device specific: what you
would filter the imagery into for a haptic device would be rather different
from what you would filter it into for creation of a tactile graphic by means
of the Tiger device.  Each of these could be implemented using SVG as a means
of expression, but as I say this is a research topic.  Reg Gollege and Bill
LaPlant have been talking about organizing [I believe an NSF] workshop to
spawn
a research program in this area.  This may not be moving fast enought for what
you are after with SVG agent round 2.  So this puts us back to being sure we
can do access to Figures #1 and #2 and taking our best guess at the
architecture that will let us absorb image understanding and haptics as growth
areas.

If there is follow-up interest I will try to make myself available.  

Al

At 11:36 PM 2001-08-05 , Jon Ferraiolo wrote:
>I am perfectly happy with your suggestion. It would be great if someone 
>researched the issue and then could share with the rest of us what he 
>discovers. This is a much more efficient process than everyone trying to 
>research everything themselves.
>
>Jon Ferraiolo
>
>At 11:32 AM 8/2/01 -0500, Jon Gunderson wrote:
>>Jon
>>I think I agree with Al [1].  I think someone like Charles should have 
>>some initial discussions with Michael Diggles to flesh out the main 
>>authoring and user agent issues.  This could then be brought back to the 
>>WAI coordination group for organizing a meeting.
>>
>>Jon
>>
>>[1]
<http://lists.w3.org/Archives/Public/w3c-wai-ua/2001JulSep/0171.html>http://
lists.w3.org/Archives/Public/w3c-wai-ua/2001JulSep/0171.html
>>
>>At 08:03 AM 8/2/2001 -0700, Jon Ferraiolo wrote:
>>>At 10:09 PM 8/1/01 -0500, Jon Gunderson wrote:
>>>>Jon,
>>>>1. Is it possible to author a WCAG compliant document with image maps
>>>>using PDF?
>>>
>>>I can't comment on the WCAG compliance question right now definitively. I 
>>>would have to read through WCAG again and do some thinking with regard to 
>>>WCAG and PDF.
>>>
>>>What exactly do you mean by image maps? Do you mean image maps in the 
>>>sense of hyperlinks on regions within image objects like HTML has? The 
>>>last I looked, PDF only supports rectangular areas for hyperlink regions, 
>>>but you can put any number of rectangular areas superimposed on images, 
>>>so you can achieve reasonably comparative results to HTML image maps, at 
>>>least from a visual rendering and operation via mouse-like pointing 
>>>device. (Obviously, lots of semantics are lost if a single hot spot in 
>>>HTML is converted into many rectangular hot spots in PDF.)
>>>
>>>
>>>>2. Is the PDF document using SVG as the internal format?
>>>
>>>Not this particular document. It is all PDF. I believe the document is 
>>>just an example of the sort of information that one might want to make 
>>>accessible. I believe we all have to use our imagination to think about 
>>>how to repackage the information to make the information more accessible.
>>>
>>>PDF does not support embedded SVG today.
>>>
>>>
>>>>3. Could you send a copy of the document to the working groups for
>>>>evaluation before the discussion?
>>>
>>>The document is accessible via the web at the URL I provided below.
>>>
>>>
>>>>4. I would be interested in particpating in the dicussion.
>>>
>>>Great.
>>>
>>>Jon Ferraiolo
>>>
>>>
>>>>Jon Gunderson
>>>>
>>>>On Wed, 1 Aug 2001, Jon Ferraiolo wrote:
>>>>
>>>> > As the SVG and WAI working groups discuss WCAG and UAAG, it can be 
>>>> helpful
>>>> > to have some sample content to evaluate against.
>>>> >
>>>> > Michael Diggles has produced a PDF file with some embedded maps and has
>>>> > issued a challenge asking how someone might approach making this 
>>>> particular
>>>> > real-world graphical content accessible. I was wondering if perhaps
>>>> > volunteers from the SVG and WAI working groups might be willing to do a
>>>> > conference call with Michael to have him explain the sorts of
information
>>>> > that his maps contain, and then have the WAI team explain the sorts of
>>>> > accessibility issues that could possible be addressed, and then review
>>>> > WCAG, UAAG and SVG to see how they measure up.
>>>> >
>>>> > Ian, it seems like this sort of a thing might be something which would
>>>> > interest you.
>>>> >
>>>> > Jon
>>>> >
>>>> > >X-Mailer: exmh version 2.3.1 01/18/2001 with version: MH 6.8.3
#10[UCI]
>>>> > >To: jferraio@adobe.com
>>>> > >Subject: Is this a Job for SVG?
>>>> > >
>>>> > >IN talking with the USGS about accessibility and PDF this morning, 
>>>> there was
>>>> > >some discussion of the Diggle Challenge. I looked at this file, and 
>>>> there
>>>> > >are,
>>>> > >no unexpectedly, some very complex figures. My question for you: is 
>>>> this the
>>>> > >sort of thing for which SVG could produce an accessible solution?
>>>> > >
>>>> > >         Loretta
>>>> > >
>>>> > >------- Forwarded Message
>>>> > >
>>>> > >From: Michael Diggles <mdiggles@usgs.gov>
>>>> > >Subject: Diggles' PDF with four figures - challenge
>>>> > >Cc: kkirk@usgs.gov
>>>> > >X-MIMETrack: Itemize by SMTP Server on 
>>>> gscamnlh01/SERVER/USGS/DOI(Release
>>>> > >5.0.7 |March
>>>> > >  21, 2001) at 07/31/2001 12:18:52 PM,
>>>> > >         Serialize by Router on gscamnlh01/SERVER/USGS/DOI(Release
5.0.7
>>>> > > |March 21,
>>>> > >2001) at
>>>> > >  07/31/2001 12:18:59 PM,
>>>> > >         Serialize complete at 07/31/2001 12:18:59 PM
>>>> > >Content-Type: text/plain; charset="us-ascii" ; format="flowed"
>>>> > >Content-Length: 2065
>>>> > >
>>>> > >Hi Loretta,
>>>> > >
>>>> > >I'm sorry I missed the conference call this morning.  I talked to
>>>> > >Keith Kirk this morning after it and he said you would like to try
>>>> > >your hand at my "challenge" tagging problem.
>>>> > >
>>>> > >Here is my note to the listserve:
>>>> > >
>>>> > > >From mkwitter@usgs.gov
>>>> > > >
>>>> > > >Is everyone out there still in the state of shock or are we the only
>>>> > > >District that relies totally on Macs to produce reports?  The 
>>>> statement at
>>>> > > >the end of the second paragraph (see web site below) of the
"Guide to
>>>> > > >making 508-compliant PDF files"..."The Macintosh operating system is
>>>> > > >incapable of tagging PDF files for accessibility."
>>>> > >
>>>> > >I don't see any evidence that Windows machines do a much better job.
>>>> > >
>>>> > >Here is a 3-foot-by 4-foot poster with four figures in it.
>>>> > >
>>>> >
><http://caldera.wr.usgs.gov/InSAR_slo.pdf>http://caldera.wr.usgs.gov/InSAR_
slo.pdf
>>>> > >
>>>> > >I will buy a beer for anybody who can make it accessible other than
>>>> > >using the strip-the-graphics technique.  I use both a Mac and a PC
>>>> > >and I gave up and made a stripped version and a graphic-containing
>>>> > >version for between now and when we figure out how to tag PDF's that
>>>> > >come from the plethora of PostScript-generating applications with
>>>> > >which we have been making our living for years.  I do admit that for
>>>> > >the version with the graphics stripped out, I moved it from my Mac to
>>>> > >my PC to use the Make-Accessible plug-in.  Just for fun, here is the
>>>> > >stripped version:  it passed the accessibility checker
>>>> > >
>>>> >
><http://caldera.wr.usgs.gov/InSAR_slo_Accessible.pdf>http://caldera.wr.usgs
.gov/InSAR_slo_Accessible.pdf
>>>> > >
>>>> > >I guess I'll put a little front-end on these two files that say
>>>> > >something like "click here for graphic-rich version" and "click here
>>>> > >for accessible version"
>>>> > >
>>>> > >I'd love to have a clean way to do it; so that offer of a beer is
open.
>>>> > >
>>>> > >- --Mike Diggles
>>>> > >~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>>>> > >Michael F. Diggles, Geologist
>>>> > >Western Publications Group, CD-ROM publications
>>>> > >Building 3, Room 3-202H
>>>> > >U.S. Geological Survey, MS-951
>>>> > >345 Middlefield Road
>>>> > >Menlo Park, CA 94025
>>>> > >
>>>> > >Phone:  (650) 329-5404
>>>> > >Fax, (650) 329-5453
>>>> > ><mailto:mdiggles@usgs.gov>mailto:mdiggles@usgs.gov
>>>> > ><http://caldera.wr.usgs.gov/>http://caldera.wr.usgs.gov
>>>> > ><http://wpg.wr.usgs.gov/>http://wpg.wr.usgs.gov/
>>>> > >~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>>>> > >
>>>> > >
>>>> > >------- End of Forwarded Message
>>>> >
>>
>>Jon Gunderson, Ph.D., ATP
>>Coordinator of Assistive Communication and Information Technology
>>Division of Rehabilitation - Education Services
>>MC-574
>>College of Applied Life Studies
>>University of Illinois at Urbana/Champaign
>>1207 S. Oak Street, Champaign, IL  61820
>>
>>Voice: (217) 244-5870
>>Fax: (217) 333-0248
>>
>>E-mail: jongund@uiuc.edu
>>
>>WWW: <http://www.staff.uiuc.edu/~jongund>http://www.staff.uiuc.edu/~jongund
>>WWW: <http://www.w3.org/wai/ua>http://www.w3.org/wai/ua
>>
>  
Received on Friday, 17 August 2001 21:55:13 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 October 2009 06:50:58 GMT