Image File and Accessibility to Visually Impaired Persons

[NickName]

"For the vision impaired, SVG offers tremendous potential for
interactive Internet mapping applications as discussed by Gardner and
Bulatov (2001).".

Now, here's an SVG file with fair/medium complexity,
http://www.carto.net/papers/svg/samples/canvas.svg, even as a sighted
person, I find its source
code albeit in xml format hard to "absorb" at the first glance, let
alone a visually impaired individual. The most important element seems
to be,
<g id="schrift" visibility="visible"... </g>

If I'm not off, it seems that key information the image conveys is as
follows:
* list of the cities
* respective location and
* spatial relations between/among them

And here are some thoughts:
If the xml for the svg is constructed in a such a fasion that
a) the above most most important elment, namely, the g with
id="schrift" in this case, is placed at the top of the file (508 parser
does not have to scan the whole file ...);
b) each city is listed from left like Bregenz, Innsbruck, Salzburg ...
c) with an attribute of positionH508 with values like "Start", "East",
"West" ...
d) another attribute of postionV508 with values like "Start", "Above",
"Below" ...

Like this:
<g id="schrift" visibility="visible" 508="true">
<text x="-2222" y="1204" class="fil2 fnt1" postionH508="Start"
postionV508="Start">Bregenz</text>
<text x="-1419" y="1702" class="fil2 fnt1" postionH508="East"
postionV508="Below">Innsbruck</text>
<text x="-143" y="1027" class="fil2 fnt1" postionH508="East"
postionV508="Above">Salzburg</text>
...
</g>

A new idea, see the above new attribute of "508" in the id of "schrift"
element. When this attribute is set to true a 508 (visually impaired)
parser would extract this element otherwise ignore. Hence, an author
of an SVG file knows and decides which elements are most important and
he/she wants to convey them to a wider audience (including visually
impaired persons).

<!-- leave x, y coordinates there as is -->

And when that becomes reality SVG would be THE future graphic format
for the widest audience possible.

What do you think?

[VK]

Re: Image File and Accessibility to Visually Impaired Persons

The rule of jurisdiction: "once in - never out" (once a precedent
created, the rest is easy)

In continuation of my post at

<http://groups.google.com/group/comp.infosystems.www.authoring.html/msg/251e36ad9fc5deef>

see

<http://dailytech.com/article.aspx?newsid=4840>

I'm personally sorry for Target (it is one of my preferred
supermarkets). But if they loose this time it means a whole new huge
source of income for web-developers. So kind of mixed sensations... :-)

[NickName]

VK wrote:

Re: Image File and Accessibility to Visually Impaired Persons

The rule of jurisdiction: "once in - never out" (once a precedent
created, the rest is easy)

Good point/observation on "once in - never out". Though the
work-around solution of adding the LongDesc attribute to an image tag
to support visually impaired persons for this situation like GIF and
JPEG/JPG files exists I would think that it's both very time consuming
and less effective to convey the meaning for a complex image file. (I
know this because I've done quite a bit of conversion to HTML document
for these types of images -- extract meaning and present them in a
simple way for this group of the unfortunate).

Now, if the standard for SVG is set or revised for that matter to
include built-in support for visually impaired, for instance, each SVG
file must have at least one element that contains 508 attribute or its
equivalent for international sake with value of TRUE (common sense
suggests that should be the most important element by authoring tool or
author? not thinking on this) and the parser -- built-in function
within a browser to support this standard, that is, without an element
of 508 value of TRUE browser would complain or the like.

Then, from organizational point of view, government agencies set policy
that at least promotes use of this emerging, more powerful image
format, would that be beneficial to many?

[VK]

The rule of jurisdiction: "once in - never out" (once a precedent

created, the rest is easy)

Good point/observation on "once in - never out".

Something easy to observe in the life :-)
Before going in further, some jurisdiction details must be cleared up.
The fact that you have use attribute 508="true" while talking about
web- accessibility in general: this fact suggests that you are talking
about the Rehabilitation Act section 508

Rehabilitation Act, section 508 is a part of equal rights thus equal
public information access rights protection. Its purpose is that any
U.S. citizen - with any income level or disabilities - could access say
Bill of Rights or the latest debates in Congress.
This way section 508 governs *federally funded programs and services*
and it has *absolutely no relation* to private commercial information
and private commercial sites.

Americans with Disabilities Act (ADA) from the other side governs the
protection for disabled people in any location: federal or private. ADA
has special section called "ADA standards for accessible design" which
is often misunderstood as if ADA defines some standards for accessible
Web design. In fact it has nothing to do with the Web: "design" here
means "design of buildings and facilities": like say you have to have
WC doors wide enough so say a person on wheelchair could go in.

What is going on now with the "Blinds vs. Target" case: they are trying
to declare any online store as a facility just like a real store on the
street - so falling under "ADA standards for accessible design".
<http://www.usdoj.gov/crt/ada/stdspdf.htm>

If they succeed (the chances are 50/50) then it will mean the biggest -
since the Internet Bubble - web-(re)design explosion across the US. And
the orders will be not from some "Sally's personal blog" but from
commercial organizations. Of course all big fish will choose bigger
nets (big studios) but it will be enough of small fish left for others.
This is why I wrote about "mixed sensations" towards this case :-)

So let's drop the Section 508 as irrelevant (with any hook available
lawyers would jump on it long ago); this way the proposed attribute
could be
....ADA="true"...
or (do not make it so US-centric)
....VImpaired="true"... for visualy impared
....HImpaired="true"... for hearing problems

Though the
work-around solution of adding the LongDesc attribute to an image tag
to support visually impaired persons for this situation like GIF and
JPEG/JPG files exists I would think that it's both very time consuming
and less effective to convey the meaning for a complex image file.

And not really helpful for image maps: it has to be description for
each linked region.

Now, if the standard for SVG is set or revised for that matter to
include built-in support for visually impaired, for instance, each SVG
file must have at least one element that contains 508 attribute or its
equivalent for international sake with value of TRUE (common sense
suggests that should be the most important element by authoring tool or
author? not thinking on this) and the parser -- built-in function
within a browser to support this standard, that is, without an element
of 508 value of TRUE browser would complain or the like.

As I pointed already, 508 is irrelevant here unless we're making an
online resource for some US federal service. I don't see how it could
be *enforced* worldwide: W3C is not an international legislative body.

Then, from organizational point of view, government agencies set policy
that at least promotes use of this emerging, more powerful image
format, would that be beneficial to many?

That could help. At the same time SVG is not "an image format" like
GIF, JPEG or PNG. It is a tool to graphical representation of XML data.
This way with data properly structured it can be much easy to simply
switch the transformer: use graphics XSLT transformer by default but
textual/aural transformer for people with disabilities.

The main question remains and I still did not get a clear answer on it:
what people with disabilities themselves would prefer: a site being
adjusted or a site being left untouched so their own helper apparatus
could go ahead?

[NickName]

Something easy to observe in the life :-)

Before going in further, some jurisdiction details must be cleared up.
The fact that you have use attribute 508="true" while talking about
web- accessibility in general: this fact suggests that you are talking
about the Rehabilitation Act section 508

Rehabilitation Act, section 508 is a part of equal rights thus equal
public information access rights protection. Its purpose is that any
U.S. citizen - with any income level or disabilities - could access say
Bill of Rights or the latest debates in Congress.
This way section 508 governs *federally funded programs and services*
and it has *absolutely no relation* to private commercial information
and private commercial sites.

Americans with Disabilities Act (ADA) from the other side governs the
protection for disabled people in any location: federal or private. ADA
has special section called "ADA standards for accessible design" which
is often misunderstood as if ADA defines some standards for accessible
Web design. In fact it has nothing to do with the Web: "design" here
means "design of buildings and facilities": like say you have to have
WC doors wide enough so say a person on wheelchair could go in.

What is going on now with the "Blinds vs. Target" case: they are trying
to declare any online store as a facility just like a real store on the
street - so falling under "ADA standards for accessible design".
<http://www.usdoj.gov/crt/ada/stdspdf.htm>

If they succeed (the chances are 50/50) then it will mean the biggest -
since the Internet Bubble - web-(re)design explosion across the US. And
the orders will be not from some "Sally's personal blog" but from
commercial organizations. Of course all big fish will choose bigger
nets (big studios) but it will be enough of small fish left for others.
This is why I wrote about "mixed sensations" towards this case :-)

OK, I'm not knowledgable aobut RA 508 and ADA and mixed them up as 508.

So let's drop the Section 508 as irrelevant (with any hook available
lawyers would jump on it long ago); this way the proposed attribute
could be
...ADA="true"...
or (do not make it so US-centric)
...VImpaired="true"... for visualy impared
...HImpaired="true"... for hearing problems

Yes,
VImpaired="true"... for visualy impared -- for image type of files
HImpaired="true"... for hearing problems -- audio files
is universal. However, I added the comments to make each super clear
and relevant.

After some more thoughts we may consider to propose it to W3C.

>

Though the
work-around solution of adding the LongDesc attribute to an image tag
to support visually impaired persons for this situation like GIF and
JPEG/JPG files exists I would think that it's both very time consuming
and less effective to convey the meaning for a complex image file.

And not really helpful for image maps: it has to be description for
each linked region.

Now, if the standard for SVG is set or revised for that matter to
include built-in support for visually impaired, for instance, each SVG
file must have at least one element that contains 508 attribute or its
equivalent for international sake with value of TRUE (common sense
suggests that should be the most important element by authoring tool or
author? not thinking on this) and the parser -- built-in function
within a browser to support this standard, that is, without an element
of 508 value of TRUE browser would complain or the like.

As I pointed already, 508 is irrelevant here unless we're making an
online resource for some US federal service. I don't see how it could
be *enforced* worldwide: W3C is not an international legislative body.

No, not "enforced", more on the 'volunteer' end from software
perspective.

>

Then, from organizational point of view, government agencies set policy
that at least promotes use of this emerging, more powerful image
format, would that be beneficial to many?

That could help. At the same time SVG is not "an image format" like
GIF, JPEG or PNG. It is a tool to graphical representation of XML data.
This way with data properly structured it can be much easy to simply
switch the transformer: use graphics XSLT transformer by default but
textual/aural transformer for people with disabilities.

I used some key part of an SVG file (XML file) to comtemplate some idea
in my original post. Go figure to explain SVG / XML to me. But have
to admit calling it "image format" is loose.

>
The main question remains and I still did not get a clear answer on it:
what people with disabilities themselves would prefer: a site being
adjusted or a site being left untouched so their own helper apparatus
could go ahead?

"a site being adjusted"? What exactly did you mean?
"untouched with ... assistive technology like screen reader ... ", is
that your second option?

[VK]

NickName wrote:

VImpaired="true"... for visualy impared -- for image type of files
HImpaired="true"... for hearing problems -- audio files
is universal. However, I added the comments to make each super clear
and relevant.

After some more thoughts we may consider to propose it to W3C.

With the chances to be even considered equal to 0.000... :-) But one
can always try.

I used some key part of an SVG file (XML file) to comtemplate some idea
in my original post. Go figure to explain SVG / XML to me. But have
to admit calling it "image format" is loose.

"a site being adjusted"? What exactly did you mean?
"untouched with ... assistive technology like screen reader ... ", is
that your second option?

I'm a bit foggy on that do I exactly want to say, so sorry for this
being reflected in the sentences I wrote.
I meant that say a blind person doubtfully just sits with the keyboard
and the mouse and navigates by regular sites just like we would do with
our monitor turned off: with all her hopes that each site (besides
federal ones) will have an aural fall-back.
I would bet that this person uses some narrator and maybe even some pad
for tactile feedback. What is really used? What web-layout changes
would improve the functioning of the client's tools? What "helpers"
would only obfuscate the functioning of these tools? That is still
terra incognita for me. And based on my current researches it is terra
incognita for W3C as well. That looks like a bunch of healthy people
decided to help to blinds: without asking blinds themselves what help
would they like :-) If you have some reputable source on the matter, I
will appreciate if you post it here.

[VK]

Re: Image File and Accessibility to Visually Impaired Persons

Just found and still reading:

<http://www.w3.org/WAI/>
<http://www.w3.org/WAI/intro/wcag.php>

Thus the proposed attribute could be "WAI" or "WCAG" (?)

[NickName]

VK wrote:

Re: Image File and Accessibility to Visually Impaired Persons

Just found and still reading:

<http://www.w3.org/WAI/>
<http://www.w3.org/WAI/intro/wcag.php>

Thus the proposed attribute could be "WAI" or "WCAG" (?)

Took a quick look and queried for "svg" with no resultset returned. It
struck me a bit strange in the sense that this "format" would seem to
be much more useful than exisitng ones. The other ALTs seems so time
consuming to produce for one thing.

[VK]

NickName wrote:

VK wrote:

Re: Image File and Accessibility to Visually Impaired Persons

Just found and still reading:

<http://www.w3.org/WAI/>
<http://www.w3.org/WAI/intro/wcag.php>

Thus the proposed attribute could be "WAI" or "WCAG" (?)

Took a quick look and queried for "svg" with no resultset returned. It
struck me a bit strange in the sense that this "format" would seem to
be much more useful than exisitng ones. The other ALTs seems so time
consuming to produce for one thing.

A bit OT: even logos to use
<http://www.w3.org/WAI/WCAG1-Conformance.html>

I'm wondering if they have anything at all about visually disabled
people?

[NickName]

VK wrote:

>...

Took a quick look and queried for "svg" with no resultset returned. It
struck me a bit strange in the sense that this "format" would seem to
be much more useful than exisitng ones. The other ALTs seems so time
consuming to produce for one thing.

A bit OT: even logos to use
<http://www.w3.org/WAI/WCAG1-Conformance.html>

Interestingly, take a very good look at this guy,
"src="http://www.w3.org/WAI/wcag1A"
If not by error, some one is thinking on something ...

>I'm wondering if they have anything at all about visually disabled

people?

What you are talking about??? The current ALT attribute for IMG tag is
for visually impaired persons. But as I said before, either ALT or
LongDesc can neither convey the full meaning of a complex image nor is
it an efficient way to achieve it if possible. Hence, the idea of
looking at the potential of SVG.

[VK]

NickName wrote:

Interestingly, take a very good look at this guy,
"src="http://www.w3.org/WAI/wcag1A"
If not by error, some one is thinking on something ...

Sorry, I did not understand you comment.

I'm wondering if they have anything at all about visually disabled

people?

What you are talking about??? The current ALT attribute for IMG tag is
for visually impaired persons.

Actually ALT attribute in IMG is a very old one and originally was
intended for non-graphical UA's and UA's with image load tourned off;
no one cared of blind people when intrudicing it ;-) But now it is
indeed re-tought as the main helper for narration software.

But as I said before, either ALT or
LongDesc can neither convey the full meaning of a complex image nor is
it an efficient way to achieve it if possible. Hence, the idea of
looking at the potential of SVG.

As I said before that would be cool to find out first how a visualy
impaired person normally conducts her browsing, what software is used
and what extra neuristic such software implements. I found nothing on
the subject so far. And without this information one can easy fall into
position of a person selling refrigerators to Eskimos.

[VK]

VK wrote:

As I said before that would be cool to find out first how a visualy
impaired person normally conducts her browsing, what software is used
and what extra neuristic such software implements. I found nothing on
the subject so far. And without this information one can easy fall into
position of a person selling refrigerators to Eskimos.

I'm reading recent posts and archives of <alt.comp.blind-users>

[NickName]

Overall comment:
Probably you did not spend quality time in reading and posting with
regard to this thread.

VK wrote:

NickName wrote:

Interestingly, take a very good look at this guy,
"src="http://www.w3.org/WAI/wcag1A"
If not by error, some one is thinking on something ...

Sorry, I did not understand you comment.

Notice, in "src="http://www.w3.org/WAI/wcag1A"
It does not end with a file EXTENSION, say, .gif, .jpeg or the like.

>

>I'm wondering if they have anything at all about visually disabled
people?

What you are talking about??? The current ALT attribute for IMG tag is
for visually impaired persons.

Actually ALT attribute in IMG is a very old one and originally was
intended for non-graphical UA's and UA's with image load tourned off;
no one cared of blind people when intrudicing it ;-) But now it is
indeed re-tought as the main helper for narration software.

It could be true the initial intent for ALT attribute is for other
types of User Agent. The important thing, imho, is what accounts.

But as I said before, either ALT or
LongDesc can neither convey the full meaning of a complex image nor is
it an efficient way to achieve it if possible. Hence, the idea of
looking at the potential of SVG.

As I said before that would be cool to find out first how a visualy
impaired person normally conducts her browsing, what software is used
and what extra neuristic such software implements. I found nothing on
the subject so far. And without this information one can easy fall into
position of a person selling refrigerators to Eskimos.

This is a bit chicken and egg situation. I don't know if you have used
AT (Assistive Technology) like Jaws. Yes, I have. And as a matter, I
tried to have a visually impaired person to access the URL for the SVG
file in my original posting with his IE, his version of IE could not
render it (while mine of IE v6.x can), my intent was, after his IE
loads the SVG file he would go to the structured XML file via Source
Code and go to the core element (for city names) with my help, and then
if you can get quite a bit of info out of it. Well, his IE sucks. I
would invite him to try it on my machine at some point though. And
when I first brought up the concept (svg) to him he did not immediately
get it (it got nothing to do with being blind. No every one is
instantly receptive to unfamiliar thing).

Another part of the idea of svg (if not mentioned already in my
original post), is that a parser would automatically extract and
logical present the element(s) (to visually impaired persons) that the
image author considers most important and tag it/them to support
visually impaired persons. Then, you see, this is not only one step
process but author's intention (for the image) is faithfully conveyed
vs. current process of ADDING attribute(s) to the IMG tag MANUALLY or
the proposed one (admittedly, for which, I haven't read much).

Now, we know that some author(s) may not give it a dam? (in other
words, would not add such an attribute WAI="True" for important
element(s) in his/her SVG file even if, say, this has been proposed and
accepted as a standard, then, one can still use the current atributes
as a remedy.

P.S. Try not to include some useless analogue like frigi and Eskimos.

[NickName]

The following URL is W3C's standard about this subject:
http://www.w3.org/TR/SVG/index.html

NickName wrote:

Overall comment:
Probably you did not spend quality time in reading and posting with
regard to this thread.

VK wrote:

NickName wrote:

Interestingly, take a very good look at this guy,
"src="http://www.w3.org/WAI/wcag1A"
If not by error, some one is thinking on something ...

Sorry, I did not understand you comment.

Notice, in "src="http://www.w3.org/WAI/wcag1A"
It does not end with a file EXTENSION, say, .gif, .jpeg or the like.

I'm wondering if they have anything at all about visually disabled
people?

What you are talking about??? The current ALT attribute for IMG tag is
for visually impaired persons.

Actually ALT attribute in IMG is a very old one and originally was
intended for non-graphical UA's and UA's with image load tourned off;
no one cared of blind people when intrudicing it ;-) But now it is
indeed re-tought as the main helper for narration software.

It could be true the initial intent for ALT attribute is for other
types of User Agent. The important thing, imho, is what accounts.

But as I said before, either ALT or
LongDesc can neither convey the full meaning of a complex image nor is
it an efficient way to achieve it if possible. Hence, the idea of
looking at the potential of SVG.

As I said before that would be cool to find out first how a visualy
impaired person normally conducts her browsing, what software is used
and what extra neuristic such software implements. I found nothing on
the subject so far. And without this information one can easy fall into
position of a person selling refrigerators to Eskimos.

This is a bit chicken and egg situation. I don't know if you have used
AT (Assistive Technology) like Jaws. Yes, I have. And as a matter, I
tried to have a visually impaired person to access the URL for the SVG
file in my original posting with his IE, his version of IE could not
render it (while mine of IE v6.x can), my intent was, after his IE
loads the SVG file he would go to the structured XML file via Source
Code and go to the core element (for city names) with my help, and then
if you can get quite a bit of info out of it. Well, his IE sucks. I
would invite him to try it on my machine at some point though. And
when I first brought up the concept (svg) to him he did not immediately
get it (it got nothing to do with being blind. No every one is
instantly receptive to unfamiliar thing).

Another part of the idea of svg (if not mentioned already in my
original post), is that a parser would automatically extract and
logical present the element(s) (to visually impaired persons) that the
image author considers most important and tag it/them to support
visually impaired persons. Then, you see, this is not only one step
process but author's intention (for the image) is faithfully conveyed
vs. current process of ADDING attribute(s) to the IMG tag MANUALLY or
the proposed one (admittedly, for which, I haven't read much).

Now, we know that some author(s) may not give it a dam? (in other
words, would not add such an attribute WAI="True" for important
element(s) in his/her SVG file even if, say, this has been proposed and
accepted as a standard, then, one can still use the current atributes
as a remedy.

P.S. Try not to include some useless analogue like frigi and Eskimos.