E-mail List Archives

Search E-mail List Archives for

Number of posts in this thread: 27 (In chronological order)

From: Rakesh P
Date: Tue, May 27 2014 12:01PM
Subject: Alternate text for images having caption adjacent
No previous message | Next message →

Dear list members,

Greetings for the day.

Many of you might have known that I write articles on various
accessibility topics, happenings and technologies at
www.maxability.co.in. Recently I have published an article on "Writing
appropriate alternate text for images" at
http://www.maxability.co.in/2014/05/writing-appropriate-alternate-text-images/.
Interestingly I got a comment that says
If alt="" is set to image having a caption describing the image
adjacent to it , how does the screen reader user know that an image is
available in the page. When alt="" is used screen reader completely
ignores the image and the user will not even know the presence of
image.
Any thoughts on this will help me in correcting the article and
provide the best possible information to the readers.
Feel free to comment as reply to this email or to understand the
context have a look and comment at blog directly
http://www.maxability.co.in/2014/05/writing-appropriate-alternate-text-images/

Thanks in advance.
Rakesh
www.maxability.co.in

From: Don Mauck
Date: Tue, May 27 2014 12:06PM
Subject: Re: Alternate text for images having caption adjacent
← Previous message | Next message →

From my perspective, unless the image has an importance behind it, if it's hidden from me as a screen user, I don't care. The only images that I need to hear about would half to have a purpose.

From: Jonathan Avila
Date: Tue, May 27 2014 12:12PM
Subject: Re: Alternate text for images having caption adjacent
← Previous message | Next message →

[Rakesh wrote] Feel free to comment as reply to this email or to
understand the context have a look and comment at blog

The WCAG defines text alternatives as

Text that is programmatically associated with non-text content or
referred to from text that is programmatically associated with non-text
content. Programmatically associated text is text whose location
can be programmatically determined from the non-text content.
Example: An image of a chart is described in text in the paragraph
after the chart. The short text alternative for the chart indicates
that a description follows.
(http://www.w3.org/TR/UNDERSTANDING-WCAG20/text-equiv-all.html)

Thus, according to this definition the in page equivalent must be
associated by a short description in the alt text of the image letting the
user know where the alternative is located. Basically there needs to be
an programmatic or textual association between the alternative and the
image.

Jonathan

From: Jared Smith
Date: Tue, May 27 2014 12:25PM
Subject: Re: Alternate text for images having caption adjacent
← Previous message | Next message →

If you use alt="", the screen reader user will not generally be aware
that the image exists. But the purpose of alternative text is to
convey the content and function of an image, not indicate that an
image exists. In general, users won't really care whether the content
is conveyed via an image or via text - they just want the content.
(This is essentially what Don just wrote.)

One question that is sometimes good to ask is, "If I couldn't use an
image here, what text would I put in it's place?" This can help you
determine good alternative text. If it's important that the user know
that an image is present (this is not usually the case), then you
could indicate this in the alt attribute.

Jared

From: Jared Smith
Date: Tue, May 27 2014 12:41PM
Subject: Re: Alternate text for images having caption adjacent
← Previous message | Next message →

On Tue, May 27, 2014 at 12:12 PM, Jonathan Avila wrote:

> Thus, according to this definition the in page equivalent must be
> associated by a short description in the alt text of the image letting the
> user know where the alternative is located.

To clarify, this only applies if the alternative text is being
conveyed in context with the image, not if it is conveyed in the alt
attribute. If it's in the alt attribute, then it's programmatically
associated to the image.

But, I think this requirement often results in less-than-optimal
accessibility. Consider a photo of me in a web page with text
immediately below it of "Jared Smith". Based on this definition, one
would have to either provide alternative text that is redundant to the
actual text, or do something like this:
<img alt="Image with description below" src="jared.jpg">
Jared Smith

This is certainly not useful or efficient. The content of the image is
that it's me, Jared Smith, so why require verbose or unnecessary text
references to that content from the image itself? Consider the
overhead of dozens of these types of alt attribute references
(alt="Product image. Product title is below."???) for every product
image on a shopping web site.

Additionally, this results in an alt attribute value that is not at
all an alternative to the image itself, but is a description of the
location of the actual alternative text. This is certainly a violation
of at least the spirit of the HTML specification's definition of the
alt attribute.

If WCAG considers a "click here" link to be considered
programmatically associated to the element that contains it, then
certainly text immediately adjacent to an image could be considered
programmatically associated to that image, no?

Of course this issue can at least partially be addressed by
<figure>/<figcaption>, though I'm not aware of any AT that actually
makes this programmatic association.

Jared Smith
WebAIM.org

From: Clark, Michelle - NRCS, Washington, DC
Date: Tue, May 27 2014 12:50PM
Subject: Re: Alternate text for images having caption adjacent
← Previous message | Next message →

I further would like to add to the discussion that I took a training class earlier this year where to and not add alt text was addressed. If the image has no value other then decorative, there is no purpose to a person using screen reading technology.

As a screen reader user, I do not have a need to have superfluous comments added that have no value to me.

Michelle

From: Jonathan Avila
Date: Tue, May 27 2014 1:09PM
Subject: Re: Alternate text for images having caption adjacent
← Previous message | Next message →

[Jared wrote] If WCAG considers a "click here" link to be considered
programmatically associated to the element that contains it, then
certainly text immediately adjacent to an image could be considered
programmatically associated to that image, no?

I agree. Another point that is related is the difference between captions
and alternative text. Alt text may often be different from a caption.
Alt text should provide a replacement for the image while a caption is
generally used to indicate how the image is related to the content. For
example, a caption might be "Washington crossing the Delaware", but the
alt text for the picture would in most situations need to describe more
than that to be a replacement of the image for someone who could not see
it.

Jonathan

From: Jared Smith
Date: Tue, May 27 2014 1:33PM
Subject: Re: Alternate text for images having caption adjacent
← Previous message | Next message →

On Tue, May 27, 2014 at 1:09 PM, Jonathan Avila wrote:

> Alt text may often be different from a caption.

Indeed. And in this case, the image would need an alt attribute value
in addition to the caption.

> For example, a caption might be "Washington crossing the Delaware", but the
> alt text for the picture would in most situations need to describe more
> than that to be a replacement of the image for someone who could not see
> it.

Yep, but if the full alternative *is* in context, adding
alt="Painting. The description is in the text below." is silly and
doesn't resemble alternative text at all, but it is what the WCAG
definition and techniques suggest is necessary for conformance.

Jared

From: Whitney Quesenbery
Date: Wed, May 28 2014 7:44AM
Subject: Re: Alternate text for images having caption adjacent
← Previous message | Next message →

I'd like to push the discussion of captions v. alt text a little further.

I work on a UX magazine, where we try to take accessibility seriously and
regularly publish articles on the topic. Many of our articles include
illustrations that are screenshots or images of other design artifacts. All
of our illustrations have captions, and a direct reference in the text.

The work of making the captions and alt text work together is one of our
ongoing discussions. I'd love any input, beyond the usual rules for alt
text, as I'm working on a style guide for our authors and editors.

1. If the *content* of the screenshot is not very important, do not
transcribe the text. The alt text briefly adds any descriptive details that
help someone visualize the image and make sense of the example.

If the *content* is important, use the alt text to identify the content
(say, repeat the title of the form or content being illustrated). Include a
box at the end of the article with the full text. (And even here,
transcribe only the part of the screen that is important to illustrating
the point.) Add a link in the caption (where it's possible technically)
that points to the long description in the box.

2. If the screenshot is compared to another, and well described in caption
or text, the alt text says whether it's the "before" or "after" (or
good/bad) image, in case a reader wants to know which is which.

If the comparison is sensory, echo words from the article to characterize
the visual impact. (For example: "Screen showing bland clip art" and
"Screen showing vibrant images")


3. Avoid de-minimus descriptions like "photo" in favor of what it's a photo
of. We use a word to describe the image only if the visual style matters.
(For example "Hand-drawn sketch" or "Visual mock-up" or "Screenshot".)

4. Think about how the alt text and caption read together, trying to make
them work as a coherent information unit

Don't hide any generally meaningful information in the alt text. Put that
in the caption, even if it makes a slightly longer caption.

Our alt text is sometimes longer than usual suggestions when the extra
words are helpful in understanding the design


Whitney





On Tue, May 27, 2014 at 3:33 PM, Jared Smith < = EMAIL ADDRESS REMOVED = > wrote:

> On Tue, May 27, 2014 at 1:09 PM, Jonathan Avila wrote:
>
> > Alt text may often be different from a caption.
>
> Indeed. And in this case, the image would need an alt attribute value
> in addition to the caption.
>
> > For example, a caption might be "Washington crossing the Delaware", but
> the
> > alt text for the picture would in most situations need to describe more
> > than that to be a replacement of the image for someone who could not see
> > it.
>
> Yep, but if the full alternative *is* in context, adding
> alt="Painting. The description is in the text below." is silly and
> doesn't resemble alternative text at all, but it is what the WCAG
> definition and techniques suggest is necessary for conformance.
>
> Jared
> > > >

From: Rakesh P
Date: Thu, May 29 2014 11:57AM
Subject: Re: Alternate text for images having caption adjacent
← Previous message | Next message →

<Jared wrote:
But, I think this requirement often results in less-than-optimal
accessibility. Consider a photo of me in a web page with text
immediately below it of "Jared Smith". Based on this definition, one
would have to either provide alternative text that is redundant to the
actual text, or do something like this:

Jared Smith

This is certainly not useful or efficient. The content of the image is
that it's me, Jared Smith, so why require verbose or unnecessary text
references to that content from the image itself? Consider the
overhead of dozens of these types of alt attribute references
(alt="Product image. Product title is below."???) for every product
image on a shopping web site.>

I totally agree with you Jared in regard with verbos, but my concern is
Consider If alt="" is used instead of "Image of or photo of Jared" in
alternate text, screen reader will not announce that an image of Jared
is available in the web page. If I want to ask my sited friend to
explain how Jared looks. Tall, fair, with blue eyes etc. alt="" will
not allow me to do so.
(I am a screen reader user just trying to put my perspective not to go
into personal details though)
Thanks & Regards
Rakesh
On 5/28/14, Whitney Quesenbery < = EMAIL ADDRESS REMOVED = > wrote:
> I'd like to push the discussion of captions v. alt text a little further.
>
> I work on a UX magazine, where we try to take accessibility seriously and
> regularly publish articles on the topic. Many of our articles include
> illustrations that are screenshots or images of other design artifacts. All
> of our illustrations have captions, and a direct reference in the text.
>
> The work of making the captions and alt text work together is one of our
> ongoing discussions. I'd love any input, beyond the usual rules for alt
> text, as I'm working on a style guide for our authors and editors.
>
> 1. If the *content* of the screenshot is not very important, do not
> transcribe the text. The alt text briefly adds any descriptive details that
> help someone visualize the image and make sense of the example.
>
> If the *content* is important, use the alt text to identify the content
> (say, repeat the title of the form or content being illustrated). Include a
> box at the end of the article with the full text. (And even here,
> transcribe only the part of the screen that is important to illustrating
> the point.) Add a link in the caption (where it's possible technically)
> that points to the long description in the box.
>
> 2. If the screenshot is compared to another, and well described in caption
> or text, the alt text says whether it's the "before" or "after" (or
> good/bad) image, in case a reader wants to know which is which.
>
> If the comparison is sensory, echo words from the article to characterize
> the visual impact. (For example: "Screen showing bland clip art" and
> "Screen showing vibrant images")
>
>
> 3. Avoid de-minimus descriptions like "photo" in favor of what it's a photo
> of. We use a word to describe the image only if the visual style matters.
> (For example "Hand-drawn sketch" or "Visual mock-up" or "Screenshot".)
>
> 4. Think about how the alt text and caption read together, trying to make
> them work as a coherent information unit
>
> Don't hide any generally meaningful information in the alt text. Put that
> in the caption, even if it makes a slightly longer caption.
>
> Our alt text is sometimes longer than usual suggestions when the extra
> words are helpful in understanding the design
>
>
> Whitney
>
>
>
>
>
> On Tue, May 27, 2014 at 3:33 PM, Jared Smith < = EMAIL ADDRESS REMOVED = > wrote:
>
>> On Tue, May 27, 2014 at 1:09 PM, Jonathan Avila wrote:
>>
>> > Alt text may often be different from a caption.
>>
>> Indeed. And in this case, the image would need an alt attribute value
>> in addition to the caption.
>>
>> > For example, a caption might be "Washington crossing the Delaware", but
>> the
>> > alt text for the picture would in most situations need to describe more
>> > than that to be a replacement of the image for someone who could not
>> > see
>> > it.
>>
>> Yep, but if the full alternative *is* in context, adding
>> alt="Painting. The description is in the text below." is silly and
>> doesn't resemble alternative text at all, but it is what the WCAG
>> definition and techniques suggest is necessary for conformance.
>>
>> Jared
>> >> >> >>
> > > >

From: Jonathan Avila
Date: Thu, May 29 2014 1:23PM
Subject: Re: Alternate text for images having caption adjacent
← Previous message | Next message →

> I totally agree with you Jared in regard with verbos, but my concern is
Consider If alt="" is used instead of "Image of or photo of Jared" in
alternate text, screen reader will not announce that an image of Jared is
available in the

Yes, this does get tricky. It really depends on the situation. For
example, a caption of me at a tweetup might state "Jon at the CSUN
tweetup". But the alt text might explain a lot more detail such as the
funny hat I'm wearing or the Got 508? T-shirt I'm wearing, etc. What the
image is conveying may very widely. For example, if my profile picture
shows me hiking as sighted person could infer I like or have been hiking
and a simple caption of "Jon Avila" wouldn't tell the user is blind and
visually impaired that information -- such information could be useful in
networking, etc.

Jonathan

From: Michael Moore
Date: Thu, May 29 2014 2:45PM
Subject: Re: Alternate text for images having caption adjacent
← Previous message | Next message →

Johathan wrote

For
example, a caption of me at a tweetup might state "Jon at the CSUN
tweetup". But the alt text might explain a lot more detail such as the
funny hat I'm wearing or the Got 508? T-shirt I'm wearing, etc.

My approach (right or wrong) has been to approach the caption and the alt
text as a pair. The caption provides information about the image that
everyone needs while the alternative text provides information that anyone
who cannot see the image needs. Thus to use a portrait image as part of a
bio as an example, I would argue that the caption is unnecessary but the
alternative text is important. alt="Mike wearing a propeller beanie" or
alt="Jon with a white cane"


On Thu, May 29, 2014 at 2:23 PM, Jonathan Avila < = EMAIL ADDRESS REMOVED = >
wrote:

> > I totally agree with you Jared in regard with verbos, but my concern is
> Consider If alt="" is used instead of "Image of or photo of Jared" in
> alternate text, screen reader will not announce that an image of Jared is
> available in the
>
> Yes, this does get tricky. It really depends on the situation. For
> example, a caption of me at a tweetup might state "Jon at the CSUN
> tweetup". But the alt text might explain a lot more detail such as the
> funny hat I'm wearing or the Got 508? T-shirt I'm wearing, etc. What the
> image is conveying may very widely. For example, if my profile picture
> shows me hiking as sighted person could infer I like or have been hiking
> and a simple caption of "Jon Avila" wouldn't tell the user is blind and
> visually impaired that information -- such information could be useful in
> networking, etc.
>
> Jonathan
>
>

From: Jared Smith
Date: Thu, May 29 2014 3:22PM
Subject: Re: Alternate text for images having caption adjacent
← Previous message | Next message →

On Thu, May 29, 2014 at 11:57 AM, Rakesh P wrote:

> Consider If alt="" is used instead of "Image of or photo of Jared" in
> alternate text, screen reader will not announce that an image of Jared
> is available in the web page.

Correct, this is the intention of alt="". The fact that it's a photo
of me as opposed to text of "Jared Smith" is not really important.

> If I want to ask my sited friend to
> explain how Jared looks. Tall, fair, with blue eyes etc. alt="" will
> not allow me to do so.

You forgot dashingly handsome. And rather muscular. And charmingly
witty. ;-) And.... well, this is precisely the problem when one starts
to believe that the alt attribute should convey a description of what
the image looks like - there is absolutely no end to what could be
described. There's no way to quantify whether the text description is
equivalent to the visual presentation. It creates a mentality that
better equivalence must mean more verbose descriptions, when the
result is generally just a waste of screen reader user time. It's
important to remember that screen reader don't have a choice to listen
to the description or not - if you put it in the alt attribute, the
screen reader WILL read it when they encounter the image.

The alt attribute if NOT for describing what an image looks like!
Instead, it's for conveying a text alternative to the *content*
provided by that image. What I look like is, in most cases, not useful
or important content. If I chose not to use the image, I certainly
wouldn't be conveying in text what I look like, right? Then why force
such unnecessary information just onto screen reader users? If it's
not alternative text, it shouldn't be placed in the alt attribute.

Jared

From: Chagnon | PubCom
Date: Thu, May 29 2014 4:43PM
Subject: Re: Alternate text for images having caption adjacent
← Previous message | Next message →

Here's a monkey wrench I'm throwing into the discussion about Alt-text and
captions.

As a professional writer, editor, magazine managing editor (which means I
say what goes where, what gets said, what's important/what isn't), graphic
designer, magazine art director, and instructor for publishing, I am stating
that captions most times don't say much about the photo they're next to.
That's not their purpose in commercial periodicals and their companion
websites.

Example: National Geographic Magazine, one of the most successful
publications of any genre in the World for 125 years. With such a legacy,
you have to admit that they know what they're doing when it comes to
communication.

NatGeo Mag provided the first studies of how humans (sighted ones) respond
to design and information. What has been proven many times over the past 100
years is that a publication is "read" (please note the quote marks) in this
order:
1) The graphics are viewed. Graphics include photos, artist illustrations,
statistical graphs, maps, and logos.
2) The largest, boldest text with the most white space is read.
3) The captions next to the graphics are then read.
4) Any decorative text is read next, such as pull quotes, deck heads, kicker
heads, sidebar headings.
(Note that I still haven't listed the actual body text of the story yet.)
5) If after all these items are read and have done their "job" of hooking
the reader, then -- and only then -- is the body text or story read.

This pattern is designed into every magazine, newspaper, newsletter, and
periodical, not just National Geographic. It's also been adapted for use in
website design -- Graphics first that convey the bulk of the message, then
the text.

NatGeo has perfected this to the ultimate degree. Their captions tell the
story, too, not just the body text story. Here's an example from the
magazine's article, Brunelleshi's Dome, which describes the building of the
dome above the main cathedral of Florence, Italy. It was an architectural
feat at the time. I don't have a copy of the printed magazine with me, but a
sample from the article's webpage follows NatGeo's format.

The caption under the graphic states: "Florence began to build a new
cathedral in 1296. Wars, politics, and plague slowed construction so work on
the dome didn't commence for more than a century."

And the caption's graphic is a close-up schematic of the cathedral's dome
with a cut-away showing the structural support beams.

Two pieces of information are conveyed in the NatGeo combo of graphic and
caption: the graphic itself is carrying an awful lot of information about
the topic that often isn't mentioned in the caption or the body text. It's
telling one part of the story. The caption is telling another part of the
story and might not refer to the graphic in any way whatsoever.

In a previous example of Jared's supposed photo, a well-trained publishing
of writer/editor/designer would create this for his photo and caption
combination:

Photo: Jared at a podium presenting at a conference. PowerPoint is up on the
presentation screen behind him. A drink with a little pink paper umbrella is
in the glass.

Caption: Jared is the Associate Director of Whosies, a nonprofit
organization for the protection of whosies in cyberspace. His notable work
on whosies, their propagation, and their protection is considered the
definitive reference in the whosie industry worldwide. In 2016, Jared was
awarded the Nobel Prize in Whosie Whatziz, recognizing his outstanding
contribution to the world of whosies. Jared presented his recent thesis,
"Whosies on the Edge," at the World Symposium on Whosie Whose in Tahiti this
past December.

Given that the majority of published information is of this nature (the
other types are SMT (science-medical-technical) and advertising), how can we
create WCAG standards that allow these types of periodicals and websites to
publish as they need to like above?

Right now, publishers think that the current WCAG standards and guidelines
are telling them not to communication in this highly effective and
profitable way. Consequently, my publishing clients are balking at making
their publications accessible.

How can we create guidelines and standards that allow for the flexibility
needed by different types of published work? We need standards that work,
not put people into a straightjacket.

--Bevi Chagnon
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
- - - - - - - - - - - - - - -
www.PubCom.com - Trainers, Consultants, Designers, Developers.
Print, Web, Acrobat, XML, eBooks, and U.S. Federal Section 508
Accessibility.
Taka a Sec. 508 Class in 2014 - www.Pubcom.com/classes

From: Julie Romanowski
Date: Fri, May 30 2014 5:06AM
Subject: Re: Alternate text for images having caption adjacent
← Previous message | Next message →

Bevi, you've made some excellent points. I had a designer try to tell me this before, but not in such detail, and I'm ashamed to say that I brushed him off. I need to touch base with him again and revisit this conversation.

Oh, and Jared, I would love to read your recent thesis, "Whosies on the Edge". Sounds intriguing! ;)

From: lists
Date: Fri, May 30 2014 10:15AM
Subject: Re: Alternate text for images having caption adjacent
← Previous message | Next message →

All, I read the the thread for most part but not completely. So apologise if this suggestion has already made.

How about we provide alt ="" and add an off screen message "image" along with description?

Regards,
Srinivasu Chakravarthula | PayPal | @csrinivasu
Sent from my iPhone 5C

> On 30-May-2014, at 2:52, Jared Smith < = EMAIL ADDRESS REMOVED = > wrote:
>
>> On Thu, May 29, 2014 at 11:57 AM, Rakesh P wrote:
>>
>> Consider If alt="" is used instead of "Image of or photo of Jared" in
>> alternate text, screen reader will not announce that an image of Jared
>> is available in the web page.
>
> Correct, this is the intention of alt="". The fact that it's a photo
> of me as opposed to text of "Jared Smith" is not really important.
>
>> If I want to ask my sited friend to
>> explain how Jared looks. Tall, fair, with blue eyes etc. alt="" will
>> not allow me to do so.
>
> You forgot dashingly handsome. And rather muscular. And charmingly
> witty. ;-) And.... well, this is precisely the problem when one starts
> to believe that the alt attribute should convey a description of what
> the image looks like - there is absolutely no end to what could be
> described. There's no way to quantify whether the text description is
> equivalent to the visual presentation. It creates a mentality that
> better equivalence must mean more verbose descriptions, when the
> result is generally just a waste of screen reader user time. It's
> important to remember that screen reader don't have a choice to listen
> to the description or not - if you put it in the alt attribute, the
> screen reader WILL read it when they encounter the image.
>
> The alt attribute if NOT for describing what an image looks like!
> Instead, it's for conveying a text alternative to the *content*
> provided by that image. What I look like is, in most cases, not useful
> or important content. If I chose not to use the image, I certainly
> wouldn't be conveying in text what I look like, right? Then why force
> such unnecessary information just onto screen reader users? If it's
> not alternative text, it shouldn't be placed in the alt attribute.
>
> Jared
> > >

From: Jared Smith
Date: Fri, May 30 2014 10:40AM
Subject: Re: Alternate text for images having caption adjacent
← Previous message | Next message →

On Fri, May 30, 2014 at 10:15 AM, = EMAIL ADDRESS REMOVED = wrote:

> How about we provide alt ="" and add an off screen message "image" along with description?

There is no accessibility requirement that you provide a description
of what the image looks like and there's no requirement that users
know that an image is present. All that is required is that you
provide alternative content for the image. In short, image
descriptions (what the image looks like) are usually not the same as
alternative text (the content the image conveys).

Alternative text is required for accessibility and goes in the alt
attribute or in adjacent text. Image descriptions or identifications
are NOT required for accessibility and should usually NOT go in the
alt attribute (unless the description and alternative text happen to
be the same). If you choose to add verbose descriptions of the image
appearance (as Bevi described regarding NatGeo Magazine examples),
they should probably be made available to everyone in a caption.

When we train about alternative text, we never use the word
"description" - instead we use "content" and "function". When you
focus on providing the image content and function in alternative text
(whether via the alt attribute or adjacent text), then the results are
usually better, more succinct equivalents.

Jared

From: Rakesh P
Date: Fri, May 30 2014 11:04AM
Subject: Re: Alternate text for images having caption adjacent
← Previous message | Next message →

On 5/30/14, Jonathan Avila < = EMAIL ADDRESS REMOVED = > wrote:
>> I totally agree with you Jared in regard with verbos, but my concern is
> Consider If alt="" is used instead of "Image of or photo of Jared" in
> alternate text, screen reader will not announce that an image of Jared is
> available in the
>
> Yes, this does get tricky. It really depends on the situation. For
> example, a caption of me at a tweetup might state "Jon at the CSUN
> tweetup". But the alt text might explain a lot more detail such as the
> funny hat I'm wearing or the Got 508? T-shirt I'm wearing, etc. What the
> image is conveying may very widely. For example, if my profile picture
> shows me hiking as sighted person could infer I like or have been hiking
> and a simple caption of "Jon Avila" wouldn't tell the user is blind and
> visually impaired that information -- such information could be useful in
> networking, etc.
>
> Jonathan
>
>

From: Rakesh P
Date: Fri, May 30 2014 11:13AM
Subject: Re: Alternate text for images having caption adjacent
← Previous message | Next message →

Thanks all for your suggestions. I have come to half-hearted
conclusion as follows.
1. If caption is available after the image and content follows the
caption, I will leave image with alt="". I will recommend the word
photo , screenshot etc along with the caption atleast as offscreen
text.
2. If caption is not available after the image , I will recommend
alternate text for the image, and the description of the image
follows it.
In this way we can reduce the verbos for screen reader user, and in
both the cases user will know the availability of image.

Thanks & Regards
Rakesh
On 5/30/14, Rakesh P < = EMAIL ADDRESS REMOVED = > wrote:
> On 5/30/14, Jonathan Avila < = EMAIL ADDRESS REMOVED = > wrote:
>>> I totally agree with you Jared in regard with verbos, but my concern is
>> Consider If alt="" is used instead of "Image of or photo of Jared" in
>> alternate text, screen reader will not announce that an image of Jared is
>> available in the
>>
>> Yes, this does get tricky. It really depends on the situation. For
>> example, a caption of me at a tweetup might state "Jon at the CSUN
>> tweetup". But the alt text might explain a lot more detail such as the
>> funny hat I'm wearing or the Got 508? T-shirt I'm wearing, etc. What the
>> image is conveying may very widely. For example, if my profile picture
>> shows me hiking as sighted person could infer I like or have been hiking
>> and a simple caption of "Jon Avila" wouldn't tell the user is blind and
>> visually impaired that information -- such information could be useful in
>> networking, etc.
>>
>> Jonathan
>>
>>

From: Whitney Quesenbery
Date: Fri, May 30 2014 2:12PM
Subject: Re: Alternate text for images having caption adjacent
← Previous message | Next message →

Bevi,

Nicely said.

And, you do something else that I think standards should do better: start
from what we actually know about how things work, from the experts who do
it.




On Thu, May 29, 2014 at 6:43 PM, Chagnon | PubCom < = EMAIL ADDRESS REMOVED = >
wrote:

> Here's a monkey wrench I'm throwing into the discussion about Alt-text and
> captions.
>
> As a professional writer, editor, magazine managing editor (which means I
> say what goes where, what gets said, what's important/what isn't), graphic
> designer, magazine art director, and instructor for publishing, I am
> stating
> that captions most times don't say much about the photo they're next to.
> That's not their purpose in commercial periodicals and their companion
> websites.
>
> Example: National Geographic Magazine, one of the most successful
> publications of any genre in the World for 125 years. With such a legacy,
> you have to admit that they know what they're doing when it comes to
> communication.
>
> NatGeo Mag provided the first studies of how humans (sighted ones) respond
> to design and information. What has been proven many times over the past
> 100
> years is that a publication is "read" (please note the quote marks) in this
> order:
> 1) The graphics are viewed. Graphics include photos, artist illustrations,
> statistical graphs, maps, and logos.
> 2) The largest, boldest text with the most white space is read.
> 3) The captions next to the graphics are then read.
> 4) Any decorative text is read next, such as pull quotes, deck heads,
> kicker
> heads, sidebar headings.
> (Note that I still haven't listed the actual body text of the story yet.)
> 5) If after all these items are read and have done their "job" of hooking
> the reader, then -- and only then -- is the body text or story read.
>
> This pattern is designed into every magazine, newspaper, newsletter, and
> periodical, not just National Geographic. It's also been adapted for use in
> website design -- Graphics first that convey the bulk of the message, then
> the text.
>
> NatGeo has perfected this to the ultimate degree. Their captions tell the
> story, too, not just the body text story. Here's an example from the
> magazine's article, Brunelleshi's Dome, which describes the building of the
> dome above the main cathedral of Florence, Italy. It was an architectural
> feat at the time. I don't have a copy of the printed magazine with me, but
> a
> sample from the article's webpage follows NatGeo's format.
>
> The caption under the graphic states: "Florence began to build a new
> cathedral in 1296. Wars, politics, and plague slowed construction so work
> on
> the dome didn't commence for more than a century."
>
> And the caption's graphic is a close-up schematic of the cathedral's dome
> with a cut-away showing the structural support beams.
>
> Two pieces of information are conveyed in the NatGeo combo of graphic and
> caption: the graphic itself is carrying an awful lot of information about
> the topic that often isn't mentioned in the caption or the body text. It's
> telling one part of the story. The caption is telling another part of the
> story and might not refer to the graphic in any way whatsoever.
>
> In a previous example of Jared's supposed photo, a well-trained publishing
> of writer/editor/designer would create this for his photo and caption
> combination:
>
> Photo: Jared at a podium presenting at a conference. PowerPoint is up on
> the
> presentation screen behind him. A drink with a little pink paper umbrella
> is
> in the glass.
>
> Caption: Jared is the Associate Director of Whosies, a nonprofit
> organization for the protection of whosies in cyberspace. His notable work
> on whosies, their propagation, and their protection is considered the
> definitive reference in the whosie industry worldwide. In 2016, Jared was
> awarded the Nobel Prize in Whosie Whatziz, recognizing his outstanding
> contribution to the world of whosies. Jared presented his recent thesis,
> "Whosies on the Edge," at the World Symposium on Whosie Whose in Tahiti
> this
> past December.
>
> Given that the majority of published information is of this nature (the
> other types are SMT (science-medical-technical) and advertising), how can
> we
> create WCAG standards that allow these types of periodicals and websites to
> publish as they need to like above?
>
> Right now, publishers think that the current WCAG standards and guidelines
> are telling them not to communication in this highly effective and
> profitable way. Consequently, my publishing clients are balking at making
> their publications accessible.
>
> How can we create guidelines and standards that allow for the flexibility
> needed by different types of published work? We need standards that work,
> not put people into a straightjacket.
>
> --Bevi Chagnon
> - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
> - - - - - - - - - - - - - - -
> www.PubCom.com - Trainers, Consultants, Designers, Developers.
> Print, Web, Acrobat, XML, eBooks, and U.S. Federal Section 508
> Accessibility.
> Taka a Sec. 508 Class in 2014 - www.Pubcom.com/classes
>
>
> > > >

From: Whitney Quesenbery
Date: Fri, May 30 2014 2:24PM
Subject: Re: Alternate text for images having caption adjacent
← Previous message | Next message →

I'm sorry Jared, but I don't understand the distinction you are making.

I ask the question more journalistically: what information is this image
adding to the page, or why has it been placed here.

That information might be:

* Text embedded in the image
* Sensory
* An illustration of a point made in the text
* Detail not easily described in words
* An example of something being discussed

The alt text needs to be written taking into account the other information
on the page.

* Is there a caption, and what does it say?
* Is the information in the image repeated or explained in the text?
* Is the image of value on its own?

A few more examples:

The text describes a process flow, and is accompanied by a diagram with the
same information. I might write alt text that says "Diagram of the process"
so all readers know what this image is, but not repeating any more
information.


Last, I NEVER use null alt text for images that are substantive
illustrations. I have had too many occasions when someone wanted to know
that it was available, or to be able to identify which one it is to send it
to someone.

The biggest exception in my mind are thumbnails that are really elaborated
visual bullets in a list or menu or ...

But one of the big gaps in many CMS is that their automated generators
don't let you indicate null alt, so you have to manually edit code or
markup, because they can't tell the difference between "blank" and "null".
All of this just the sort of boring problem of programs that don't consider
the authoring user interface and create tools that support, instead of
actively making it hard, to make content accessible.


On Fri, May 30, 2014 at 12:40 PM, Jared Smith < = EMAIL ADDRESS REMOVED = > wrote:

> On Fri, May 30, 2014 at 10:15 AM, = EMAIL ADDRESS REMOVED = wrote:
>
> > How about we provide alt ="" and add an off screen message "image" along
> with description?
>
> There is no accessibility requirement that you provide a description
> of what the image looks like and there's no requirement that users
> know that an image is present. All that is required is that you
> provide alternative content for the image. In short, image
> descriptions (what the image looks like) are usually not the same as
> alternative text (the content the image conveys).
>
> Alternative text is required for accessibility and goes in the alt
> attribute or in adjacent text. Image descriptions or identifications
> are NOT required for accessibility and should usually NOT go in the
> alt attribute (unless the description and alternative text happen to
> be the same). If you choose to add verbose descriptions of the image
> appearance (as Bevi described regarding NatGeo Magazine examples),
> they should probably be made available to everyone in a caption.
>
> When we train about alternative text, we never use the word
> "description" - instead we use "content" and "function". When you
> focus on providing the image content and function in alternative text
> (whether via the alt attribute or adjacent text), then the results are
> usually better, more succinct equivalents.
>
> Jared
> > > >

From: Denis Boudreau
Date: Fri, May 30 2014 3:17PM
Subject: Re: Alternate text for images having caption adjacent
← Previous message | Next message →

Hi everyone,

On May 29, 2014, at 5:22 PM, Jared Smith < = EMAIL ADDRESS REMOVED = > wrote:

> On Thu, May 29, 2014 at 11:57 AM, Rakesh P wrote:
>
>> Consider If alt="" is used instead of "Image of or photo of Jared" in
>> alternate text, screen reader will not announce that an image of Jared
>> is available in the web page.
>
> Correct, this is the intention of alt="". The fact that it's a photo
> of me as opposed to text of "Jared Smith" is not really important.


This is precisely what I have been advocating for years. That is, until @steveofmaine (https://twitter.com/steveofmaine) threw an unexpected monkey wrench into my clearly defined little world a couple of weeks ago.

Whether we choose to provide a null alt attribute value to pictures of people’s portraits on a web page all boils down to one's interpretation of what "purely decorative images” actually means. Luckily, WCAG 2.0 provides a definition of "pure decoration” that goes like this: an image "serving only an aesthetic purpose, providing no information, and having no functionality” (http://www.w3.org/TR/WCAG20/#puredecdef). Seems simple enough, right?

I’m sure most would agree that pictures of people right next to their name mostly serve aesthetic purposes. Pictures allow most of us to know what the person actually looks like, but we can’t exactly use the alt text value of an image to describe how the person physically looks. That would be awkward! Therefore, the alt text ends up repeating the person’s name and consequently, provides no real additional information that the adjacent text doesn’t already provide. Which, I assume, is why Jared was saying it’s not really important.

However, going back to the definition of “pure decoration” and looking into the concept of having no functionality, I now question whether or not a portrait of a person does, in fact, provide functionality. And I think it does. When we decide a photo should be marked up with alt=“” because the person’s name is right next to the image, we’re taking away what seems to be two functionalities from non-sighted screen reader users:

1. The ability to locate a person’s info based on the fact that it’s right next to his or her picture. Pictures, like headings or other objects in the page, provide structure. If a screen reader user has been told there’s a picture of the person on the page and they don’t know that person's name, they might want to search the page through images to find it. Assuming they don’t know what that person’s name is, searching for images and looking for text right next to it could be an efficient way to search the page. When we make the image invisible to screen reader users, then users cannot use the image anymore to quickly find the information about the person. Relying on null alt text takes that functionality or ability away.

2. The ability to actually locate the picture, and download it (for whatever purpose) is another thing we take away from these users when we choose to go with null alt text. If you can't know a picture is available on the page because authors have decided to make it invisible to you, you can’t do anything with it / about it. By assigning a null alt value to an image that in fact, does convey some kind of information (this is what this person looks like), we take away the non-sighted users’ ability to do something with that image. That, too, seems to be a functionality we’re taking away from non-sighted screen reader users.

Which is why I now believe that such photos need to have descriptive alt text, and that it has some degree of importance.

No having said that, I’m not saying those functionalities are the most important things in the world, but they’re functionalities nonetheless… and who would we be to decide whether or not users actually need those images?

I think there’s a real difference between a purely decorative image used for a border and a portrait and as such, the latter should not count as purely decorative images. I think we need to make sure portrait images do have alt text values assigned to them: it would mean a little repetition in the content, but then so be it. Personally, I’d rather risk giving a little too much to users, rather than actually taking something away from them.

/Denis



--
Denis Boudreau,
Web Accessibility Avenger
DEQUE Systems Inc.
Tel: +1 (514) 730-9168
Web: www.deque.com

From: Jared Smith
Date: Fri, May 30 2014 3:23PM
Subject: Re: Alternate text for images having caption adjacent
← Previous message | Next message →

On Fri, May 30, 2014 at 2:24 PM, Whitney Quesenbery wrote:

> I ask the question more journalistically: what information is this image
> adding to the page, or why has it been placed here.

These are certainly good questions to ask to determine the content of
the image, and thus the alternative text for that image.

> That information might be:
>
> * Text embedded in the image
> * Sensory
> * An illustration of a point made in the text
> * Detail not easily described in words
> * An example of something being discussed

If that information is determined to be useful content, then yes, it
should be conveyed.

> Last, I NEVER use null alt text for images that are substantive
> illustrations.

Likewise. It is of note that these types of highly illustrative,
journalistic images constitute a small percentage of images on the
web.

Our surveys - http://webaim.org/projects/screenreadersurvey5/ -
provide a good example of a brief alternative used on the complex
charts with the full text alternative in context. If one were to focus
on a *description* of one of the charts, it might be alt="A circular
pie chart with slices that are varying colors of red" - hardly an
equivalent, right?

As another example, imagine a photo of a woman wearing a headset in
the sidebar of a homepage. The image is linked to the customer service
area of the site. If you focus on providing a description of the
image, you'd probably use alt="smiling blonde woman wearing a headset
while sitting at a computer". Instead, alt="Customer Service" or
similar would be better and most efficient. A *description* of what
the image looks like or that it contains a smiling blonde woman is, in
this case, not relevant to the much-more-important *content* and
*function* of the image - that clicking it takes you to the customer
support page.

In our evaluation work we see thousands of images with alternative
text that simply describes what the thing looks like with little
regard to the actual content being conveyed. All I'm suggesting is
that we should rethink how we describe and present what alternative
text is so that the focus is placed on content, not visual
descriptions.

Jared

From: Jared Smith
Date: Fri, May 30 2014 3:55PM
Subject: Re: Alternate text for images having caption adjacent
← Previous message | Next message →

On Fri, May 30, 2014 at 3:17 PM, Denis Boudreau wrote:

> Luckily, WCAG 2.0 provides a definition of "pure decoration"

There's a conversation going on right now in the working group
(http://www.w3.org/WAI/GL/track/issues/21) about the confusing nature
of this definition and that it doesn't match reality or Steve's HTML5
definitions and examples at
http://dev.w3.org/html5/alt-techniques/#sec5

> we can't exactly use the alt text value of an image to describe
> how the person physically looks. That would be awkward!

Not only awkward, but it would be a misuse of the alt attribute if the
description of what they look like is not an alternative to the
content of the image.

> Therefore, the alt text ends up repeating the person's name and
> consequently, provides no real additional information that the
> adjacent text doesn't already provide. Which, I assume, is why
> Jared was saying it's not really important.

The information certainly is important. But it shouldn't be conveyed
twice. Conveying it once in the caption is sufficient.

> If a screen reader user has been told there's a picture of the
> person on the page and they don't know that person's name,
> they might want to search the page through images to find it.

If a screen reader user can't search through the images on a page,
this is a screen reader deficiency. The burden should not be placed on
authors to say, "Hey screen reader, here's an image". We certainly
have no requirement to identify forms or tables, for example, with
text that indicates they are present. Screen readers do this based on
the semantics of the page.

If an image with alt="" has a programmatic association to content
(adjacent text, aria-label or description, <figcaption>, etc.), then
assistive technologies should utilize this when presenting an image
list or while navigating images.

> Relying on null alt text takes that functionality or ability away.

No, screen reader deficiencies don't make that functionality
available. I think it's a slippery slope if we start creatively
interpreting accessibility guidelines and HTML specifications to fill
gaps in screen reader behavior.

> 2. The ability to actually locate the picture, and download it (for whatever purpose) is another thing we take away from these users when we choose to go with null alt text.

As above, this is a user agent issue, not an authoring issue.
Certainly the alt attribute is NOT defined as "an indicator of the
presence of an image" - that's what <img> is for.

I would, however, argue that if an image is of such importance that
the user would likely have a need to download it, then it probably
necessitates at least a brief alternative as Whitney and I suggested
previously.

It continues to fascinate me how complex and subject to interpretation
alternative text is despite being the foundational technique of our
entire field. Thanks for engaging in the dialogue!

Jared

From: Chagnon | PubCom
Date: Fri, May 30 2014 7:44PM
Subject: Re: Alternate text for images having caption adjacent
← Previous message | Next message →

Where are the editors when we need them?!

Jared wrote: "As another example, imagine a photo of a woman wearing a
headset in the sidebar of a homepage. The image is linked to the customer
service area of the site. ... Instead, alt="Customer Service" or similar
would be better and most efficient."

Agree, but let's also make sure our writing of Alt-text is grammatically
correct and uses clear, plain language. "Customer Service" is an intangible
action, not the human being portrayed in the photo. I'd argue that a better,
grammatically correct Alt-Text would be: "Customer service representative."

However, if the words "Customer Service" were embedded in the graphic or as
live text next to it, then the graphic itself becomes less relevant and can
be nulled or changed to "Customer Service."

But the best language to use in this example is "Contact Customer Service"
because it contains an action verb which clearly, simply presents the action
the website builder is encouraging the user to do -- in other words, the
intent.

"... much-more-important *content* and *function* of the image - that
clicking it takes you to the customer support page..."
Yes, that too. Today, that graphic is also likely to connect you to a chat
session with a customer service rep. So we better make sure our links and
alt-text states precisely when the user will be connected to a chat session
versus going to a different webpage.

What we need are professional editors who specialize in writing Alt-Text
that clearly describes the intent of the graphic, as well as making sure
that hyperlinks are clearly written and specify the destination if it's
clicked.

There are few of these editors around, but I'm working on it! Have a class
running in July just for writers and editors, and a big portion of the class
is how to write Alt-text. This is a great discussion for the class!

--Bevi Chagnon
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
- - - - - - - - - - - - - - -
www.PubCom.com - Trainers, Consultants, Designers, Developers.
Print, Web, Acrobat, XML, eBooks, and U.S. Federal Section 508
Accessibility.
Take a Sec. 508 Class in 2014 - www.Pubcom.com/classes

From: Whitney Quesenbery
Date: Fri, May 30 2014 9:43PM
Subject: Re: Alternate text for images having caption adjacent
← Previous message | Next message →

On Fri, May 30, 2014 at 5:23 PM, Jared Smith < = EMAIL ADDRESS REMOVED = > wrote:

> On Fri, May 30, 2014 at 2:24 PM, Whitney Quesenbery wrote:
>
>
> > Last, I NEVER use null alt text for images that are substantive
> > illustrations.
>
> Likewise. It is of note that these types of highly illustrative,
> journalistic images constitute a small percentage of images on the
> web.
>

If you are looking at the entire page and frame, that's probably true.
But if you are thinking about the article content - the stuff in <main>,
then it is not. And that's what most of the "web authors" out there are
focused on.



> In our evaluation work we see thousands of images with alternative
> text that simply describes what the thing looks like with little
> regard to the actual content being conveyed. All I'm suggesting is
> that we should rethink how we describe and present what alternative
> text is so that the focus is placed on content, not visual
> descriptions.


I certainly agree with this! And with the bad-ness of the examples you
shared.

One of the reasons for this is that we too often make programmers write the
alt text. That is, we say "this is code so a coder should do it" instead of
"this is part of the content, so it should be written by an author". Our
tools push us in that direction when they hide the alt text in a properties
box or with a lot of other technical entries. Or just make it hard to see.
Why, for example, can't I see a panel with all of the alt text for all of
the illustrations in my article at once, so that I can edit them (a)
consistently and (b) while I'm looking at the rest of the text - as a
whole, not as an isolated <img> tag.

You may also notice that I've avoided the technical label "image". They may
be in a graphical file format, but it can serve many purposes on a page.
The person who "owns" the purpose should also "own the alternative text"

From: Steve Faulkner
Date: Sat, May 31 2014 1:28AM
Subject: Re: Alternate text for images having caption adjacent
← Previous message | No next message

On a related subject:

HTML5: Techniques for providing useful text alternatives
http://rawgit.com/w3c/alt-techniques/master/index.html

is currently being considered for publication as a W3C Note.



--

Regards

SteveF
HTML 5.1 <http://www.w3.org/html/wg/drafts/html/master/>;


On 27 May 2014 19:01, Rakesh P < = EMAIL ADDRESS REMOVED = > wrote:

> Dear list members,
>
> Greetings for the day.
>
> Many of you might have known that I write articles on various
> accessibility topics, happenings and technologies at
> www.maxability.co.in. Recently I have published an article on "Writing
> appropriate alternate text for images" at
>
> http://www.maxability.co.in/2014/05/writing-appropriate-alternate-text-images/
> .
> Interestingly I got a comment that says
> If alt="" is set to image having a caption describing the image
> adjacent to it , how does the screen reader user know that an image is
> available in the page. When alt="" is used screen reader completely
> ignores the image and the user will not even know the presence of
> image.
> Any thoughts on this will help me in correcting the article and
> provide the best possible information to the readers.
> Feel free to comment as reply to this email or to understand the
> context have a look and comment at blog directly
>
> http://www.maxability.co.in/2014/05/writing-appropriate-alternate-text-images/
>
> Thanks in advance.
> Rakesh
> www.maxability.co.in
> > > >