alt text

N

Neredbojias

The last several months I've been trying to include more reasonable alt
text in my images to improve accessibility. The other day I uploaded a new
page and forgot to upload the matching folder containing the page's images.
Here's what I got in Firefox (look just below the line near the top which
says "Featured Images - Interesting Chairs"):

http://www.neredbojias.com/tsttst/capalt1b.jpg

Here's how it looked in Safari 3.0.4 for Windows:

http://www.neredbojias.com/tsttst/capalt2b.jpg

Don't they know how to do it right, or am I missing something? Opera and
IE7 were okay.
 
J

Jukka K. Korpela

Scripsit Neredbojias:
The last several months I've been trying to include more reasonable
alt text in my images to improve accessibility.
OK.

The other day I
uploaded a new page and forgot to upload the matching folder
containing the page's images. Here's what I got in Firefox

It's a blurred picture of something. What is the point?
Don't they know how to do it right,

Who's "they" and what's "it"?

If you're criticizing a browser for doing something wrong, as I suspect,
please specify the browser in detail (including version), the specific
URL(s) you tested, and a verbal explanation of what you regard as wrong.
 
H

Harlan Messinger

Neredbojias said:
The last several months I've been trying to include more reasonable alt
text in my images to improve accessibility. The other day I uploaded a new
page and forgot to upload the matching folder containing the page's images.
Here's what I got in Firefox (look just below the line near the top which
says "Featured Images - Interesting Chairs"):

http://www.neredbojias.com/tsttst/capalt1b.jpg

Here's how it looked in Safari 3.0.4 for Windows:

http://www.neredbojias.com/tsttst/capalt2b.jpg

Don't they know how to do it right, or am I missing something? Opera and
IE7 were okay.
"User agents must render alternate text when they cannot support images,
they cannot support a certain image type or when they are configured not
to display images."

http://www.w3.org/TR/html4/struct/objects.html#h-13.2

None of these is the case. That's not to say that the browsers couldn't
handle the case of a missing file in the same way, but the spec doesn't
call for it.

However: are you setting width and height for your images? If so, are
you sure that the ALT text isn't just being cut off because the "missing
image" icon isn't already filling up the space you made available?
 
J

Jukka K. Korpela

Scripsit Harlan Messinger:
"User agents must render alternate text when they cannot support
images, they cannot support a certain image type or when they are
configured not to display images."

http://www.w3.org/TR/html4/struct/objects.html#h-13.2

None of these is the case. That's not to say that the browsers
couldn't handle the case of a missing file in the same way, but the
spec doesn't call for it.

It does. You missed a simple statement in the section you quoted, right
before the part you quoted:

"The alt attribute specifies alternate text that is rendered when the
image cannot be displayed".
However: are you setting width and height for your images? If so, are
you sure that the ALT text isn't just being cut off because the
"missing image" icon isn't already filling up the space you made
available?

I still haven't seen a decent problem description, so I refrain from
commenting on the unknown problem.

However, at the general level, relating to the above suggestion on
solving an unknown problem: It's an age-old problem that some browsers
use the image dimensions, if specified explicitly, to determine the box
size, so that if the alt text doesn't fit, it's truncated. So for years,
it's been a common recommendation not to specify width and height for
small images with nonempty alt text.

This used to be typical of IE, and in this sad, sick world of
bug-compatible browsers, I wouldn't be surprised if Firefox started
imitating such a "feature".
 
N

Neredbojias

Scripsit Neredbojias:


It's a blurred picture of something. What is the point?


Who's "they" and what's "it"?

If you're criticizing a browser for doing something wrong, as I suspect,
please specify the browser in detail (including version), the specific
URL(s) you tested, and a verbal explanation of what you regard as wrong.

Please see my reply to your reply to Harlan Messinger in this same thread.
 
N

Neredbojias

Scripsit Harlan Messinger:


It does. You missed a simple statement in the section you quoted,
right before the part you quoted:

"The alt attribute specifies alternate text that is rendered when the
image cannot be displayed".


I still haven't seen a decent problem description, so I refrain from
commenting on the unknown problem.

However, at the general level, relating to the above suggestion on
solving an unknown problem: It's an age-old problem that some browsers
use the image dimensions, if specified explicitly, to determine the
box size, so that if the alt text doesn't fit, it's truncated. So for
years, it's been a common recommendation not to specify width and
height for small images with nonempty alt text.

That isn't the problem. The first capture, to wit:

http://www.neredbojias.com/tsttst/capalt1b.jpg

taken of Firefox 2.0.0.13 but demonstrating an anomoly existing in many
previous versions as well, shows the results of a page wherein the
thumbnail images indeed had width and height attributes included but
display alt text for missing images overlapped all on one line.

Why you asked for a url puzzles me because the images are now uploaded
and the problem cannot be seen "live", so to speak. Hence the capture.
I suppose I could upload a test page, a clone of the original with
changed, non-responsive paths, but is that really necessary?
This used to be typical of IE, and in this sad, sick world of
bug-compatible browsers, I wouldn't be surprised if Firefox started
imitating such a "feature".

In a sense, Firefox's "feature" is just the opposite.
 
J

Jukka K. Korpela

Scripsit Neredbojias:
Why you asked for a url puzzles me

It puzzles me that it puzzles you.
I suppose I could upload a test page, a clone of the
original with changed, non-responsive paths, but is that really
necessary?

Oh, only if you want to spawn a constructive discussion. Or should we
discuss problems that do not exist any more?
 
N

Neredbojias

Scripsit Neredbojias:


It puzzles me that it puzzles you.


Oh, only if you want to spawn a constructive discussion. Or should we
discuss problems that do not exist any more?

Well, I guess I can see your point. Here is a live link demonstrating the
issue:

http://www.neredbojias.com/ind23test.php

Just to pre-warn you, the alt text is inserted via javascript, so...
 
D

dorayme

Neredbojias said:
The last several months I've been trying to include more reasonable alt
text in my images to improve accessibility. The other day I uploaded a new
page and forgot to upload the matching folder containing the page's images.
Here's what I got in Firefox (look just below the line near the top which
says "Featured Images - Interesting Chairs"):

http://www.neredbojias.com/tsttst/capalt1b.jpg

Here's how it looked in Safari 3.0.4 for Windows:

http://www.neredbojias.com/tsttst/capalt2b.jpg

Don't they know how to do it right, or am I missing something? Opera and
IE7 were okay.

Ages ago, about the time I embarked on a course of alt text pills to
improve author implementations, I made a test page to see how different
browsers rendered things.

<http://tinyurl.com/34z6kf>

I thought I would wheel it out here.

Also, perhaps I should mention, but please send $10, you can order the
formula for making your own alt-text pills from me.
 
H

Harlan Messinger

Jukka said:
Scripsit Harlan Messinger:


It does. You missed a simple statement in the section you quoted, right
before the part you quoted:

"The alt attribute specifies alternate text that is rendered when the
image cannot be displayed".

I took the subsequent details to qualify the general statement, i.e.
"when the image cannot be displayed owing to client-side factors." Since
they took the trouble to enumerate three client-side factors, it seems
reasonable to take the failure to mention the obvious server-side factor
to be meaningful.
 
N

Neredbojias

Ages ago, about the time I embarked on a course of alt text pills to
improve author implementations, I made a test page to see how
different browsers rendered things.

<http://tinyurl.com/34z6kf>

I thought I would wheel it out here.

Also, perhaps I should mention, but please send $10, you can order the
formula for making your own alt-text pills from me.

Hmm, interesting. Unfortunately, it made me thirsty and now I'm too drunk
to understand it.

Seriously, what's the point and how does it relate to my particular lament?
 
D

dorayme

Neredbojias said:
what's the point and how does it relate to my particular lament?

Mine was meant as a test page for how supplied alt text appears in
different browsers under the various conditions one encounters like
image not available, alt text supplied, not supplied beyond "", not
supplied at all. You can see the parameters via the source.

But perhaps it needs better overt explanation on the web page itself
with comments like "No such image was loaded to the server" etc.

Is this something quite irrelevant to your post?
 
N

Neredbojias

Mine was meant as a test page for how supplied alt text appears in
different browsers under the various conditions one encounters like
image not available, alt text supplied, not supplied beyond "", not
supplied at all. You can see the parameters via the source.

But perhaps it needs better overt explanation on the web page itself
with comments like "No such image was loaded to the server" etc.

I think that would be very beneficial. You seem to have covered the
possibilities well but it is a little vacuous just looking at it now.
Is this something quite irrelevant to your post?

In he sense that the problem I described wouldn't even show up in your
page, yes.
 
J

Jukka K. Korpela

Scripsit Harlan Messinger:
I took the subsequent details to qualify the general statement,

That is, you omitted the statement that is closest to a _definition_ and
quoted only secondary statements after it, and now you're claiming that
they refute it:
i.e.
"when the image cannot be displayed owing to client-side factors."
Since they took the trouble to enumerate three client-side factors,
it seems reasonable to take the failure to mention the obvious
server-side factor to be meaningful.

Non sequitur. There is nothing particularly obvious in the "server-side
factor", and no particular reason why they should have listed down more
than three examples. They seldom give even that many examples.

The HTML specifications are generally sloppy, obscure, and partly
self-contradictory, and calling them "standards" is rather ridiculous.
But we have to live with them, and there's no reason to try to read even
more obscurity into them than they have.

We have every right to expect that a browser presents the alt attribute
value whenever it does not present an image specified by the src
attribute of an <img> element. Browsers don't always meet our
expectations, but in cases like this, a bug should be called a bug and
not a feature or implementation-defined thing.
 
H

Harlan Messinger

dorayme said:
It is an American thing, I understand, that "alternate" can be used as
an alternative to "alternative"

I hadn't known that but evidently so. In particular, I didn't realize
that the *noun* "alternate" used to mean "One who is appointed to act in
place of a delegate who is unable to be present; a substitute" is US
usage until looking in the OED just now.
 
D

dorayme

"Jukka K. Korpela said:
Scripsit Harlan Messinger:


That is, you omitted the statement that is closest to a _definition_ and
quoted only secondary statements after it, and now you're claiming that
they refute it:


Non sequitur. There is nothing particularly obvious in the "server-side
factor", and no particular reason why they should have listed down more
than three examples. They seldom give even that many examples.

The HTML specifications are generally sloppy, obscure, and partly
self-contradictory, and calling them "standards" is rather ridiculous.
But we have to live with them, and there's no reason to try to read even
more obscurity into them than they have.

We have every right to expect that a browser presents the alt attribute
value whenever it does not present an image specified by the src
attribute of an <img> element. Browsers don't always meet our
expectations, but in cases like this, a bug should be called a bug and
not a feature or implementation-defined thing.

13.2

"The alt attribute specifies alternate text that is rendered when the
image cannot be displayed (see below for information on how to specify
alternate text ). User agents must render alternate text when they
cannot support images, they cannot support a certain image type or when
they are configured not to display images."

The second sentence begins with a point that includes the possibility
that the image is not available due to it not being on the server at
all. A user agent naturally cannot support something that does not
exist. The writers of this paragraph were elaborating in their second
sentence. No conclusion can be drawn about the elaboration leaving out
anything.
 
T

Toby A Inkster

dorayme said:
It is an American thing, I understand, that "alternate" can be used as
an alternative to "alternative"

Indeed. Outside the US, alternate means, as a verb "to switch between
alternatives" and as an adjective it roughly means "every second" (by
"second", I mean the ordinal of two, not the unit of time).

verb e.g.: I like to alternate between ham sandwiches and cheese
sandwiches for lunch.

noun e.g.: I have ham and cheese in my sandwiches on alternate lunches.

ObHTML: I think it's quite annoying that the W3C chose "alternate" as a
link rel value rather than "alternative" which has the same meaning in
both en-GB and en-US.

--
Toby A Inkster BSc (Hons) ARCS
[Geek of HTML/SQL/Perl/PHP/Python/Apache/Linux]
[OS: Linux 2.6.17.14-mm-desktop-9mdvsmp, up 6 days, 21:46.]

Cognition 0.1 Alpha 6
http://tobyinkster.co.uk/blog/2008/03/29/cognition-alpha6/
 
H

Harlan Messinger

Toby said:
Indeed. Outside the US, alternate means, as a verb "to switch between
alternatives" and as an adjective it roughly means "every second" (by
"second", I mean the ordinal of two, not the unit of time).

verb e.g.: I like to alternate between ham sandwiches and cheese
sandwiches for lunch.

noun e.g.: I have ham and cheese in my sandwiches on alternate lunches.

That would be its use as an adjective.
ObHTML: I think it's quite annoying that the W3C chose "alternate" as a
link rel value rather than "alternative" which has the same meaning in
both en-GB and en-US.

I can understand that, yet I still find it less annoying that "REFERER"
with the missing R.
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Members online

Forum statistics

Threads
473,770
Messages
2,569,584
Members
45,076
Latest member
OrderKetoBeez

Latest Threads

Top