Re: [WSG] Web site images question

2006-06-11 Thread Patrick H. Lauke

David Dixon wrote:

Relevant to the content? From your own list of resources the alt 
attribute should be a textual alternative for the meaning of the 
image. It has no more relevance to the content than the image itself, 
and as the image's purpose is to show the user that the wheelchair is a 
symbol for accessibility (with further advisory explanation from the 
title element), then I believe the above example is perfectly valid.


It would be true if this was a page explaining symbols. As it stands, 
though, the purpose of those images is purely an aesthetic enhancement 
to accompany the text. They serve no purpose.


But whatever. It's obvious that until you ascribe meaning to those 
images, we won't agree on this.


P
--
Patrick H. Lauke
__
re·dux (adj.): brought back; returned. used postpositively
[latin : re-, re- + dux, leader; see duke.]
www.splintered.co.uk | www.photographia.co.uk
http://redux.deviantart.com
__
Web Standards Project (WaSP) Accessibility Task Force
http://webstandards.org/
__


**
The discussion list for  http://webstandardsgroup.org/

See http://webstandardsgroup.org/mail/guidelines.cfm
for some hints on posting to the list  getting help
**



Re: [WSG] Web site images question

2006-06-11 Thread John S. Britsios

David Dixon wrote:
img src=/images/accessibility.jpg width=100 height=89 
alt=The imagery of a person on a wheelchair is generally considered 
a symbol for accessibility title=An image of a wheelchair: the 
symbol for accessibility


How is that alt text *relevant* to the content at all?


Relevant to the content? From your own list of resources the alt 
attribute should be a textual alternative for the meaning of the 
image. It has no more relevance to the content than the image itself, 
and as the image's purpose is to show the user that the wheelchair is 
a symbol for accessibility (with further advisory explanation from the 
title element), then I believe the above example is perfectly valid.
But isn't the alt text attribute of the above example too long? 91 
characters inculding spaces? I always take care not have more that 60 
characters including spaces, and in the worse case no more than 70.


Ive even run tests against JAWS and the nice FF extension Fangs to 
ascertain the behaviour of the screen reader for this particular 
example, and as neither gives any indication of the advisory title 
text, then I believe the alt attribute should be giving the same 
message as the full image (ie the image plus the advisory 
information). Of course this is personal preference, but as Ive stated 
earlier, ensuring a page is accessible should be trying to make the 
page as usable and understandable for a less-abled person as for an 
abled person.
I was just wondering, if novice Internet users would really understand 
the meaning of the graphics, and I thought it would be necessary using 
at least the alt tag. Or?


img src=mycomputer alt=An image of a computer title=Shows the 
disk drives and hardware connected to this computer My Computer


img src= alt=The image of a folder and magnifying glass is the 
symbol for Windows Explorer title=Displays the files and folders on 
your computer. Windows Explorer


img src=word alt=The Microsoft Word icon is a blue W inside a 
square title=Create and edit text and graphics in letters, reports, 
Web pages, or e-mail messages by using Microsoft Office Word. 
Microsoft Word


I dont find these examples relevant to the context of my example. For 
my example, the image had a context for which is was designed... the 
content regarding accessibility. These examples of desktop icons, have 
no such context. They are their own context, in addition, their usage 
is vastly different, in a web type example, these would be links, with 
the image itself being a graphical representation of the text beside 
it, much like my earlier example of using the accessibility logo as a 
background to a header. As with that previous example, I would mark up 
the image differently than the image in this example.


Also, as for Patrick's nth degree explanation, I dont see the point 
in explaining every single aspect of the image itself to verbatim. 
Enough to explain the purpose of the image itself is quite sufficient. 
Also, my alt text does not describe the image itself (ie i dont say 
this is a symbol of a yellow person in a yellow wheelchair), but 
saying what the image describes a person in a wheelchair is a symbol 
for accessibility gives the same meaning as the image itself.
As for my cigarette example, then yes, I think giving a hint as to 
colour of the symbol is valid, as this symbol is universal (at least 
in the UK). The red circle itself symbolises something which is not 
permitted. If you were to explain what a no smoking symbol looked like 
without saying the circle was red (even if they have no concept of red 
looks like) would give a lesser clue as to its purpose as providing 
its colour. In fact, just saying a red circle, would probably give the 
idea of something that is not permitted, even before saying it has a 
picture of a cigarette in it.

Good points David.


Thanks,

David.


**
The discussion list for  http://webstandardsgroup.org/

See http://webstandardsgroup.org/mail/guidelines.cfm
for some hints on posting to the list  getting help
**





--
John S. Britsios
Web Architect  Marketing Consultant

Webnauts Net (Main Office)
Koblenzer Str. 37A
D-33613 Bielefeld

Webnauts Net (U.S. Office)
5 Ivanhoe Drive
Urbana IL 61802

Email: [EMAIL PROTECTED]
Web address: http://www.webnauts.net



**
The discussion list for  http://webstandardsgroup.org/

See http://webstandardsgroup.org/mail/guidelines.cfm
for some hints on posting to the list  getting help
**



Re: [WSG] Web site images question

2006-06-11 Thread Patrick H. Lauke

David Dixon wrote:

As for my cigarette example, then yes, I think giving a hint as to 
colour of the symbol is valid, as this symbol is universal (at least in 
the UK). The red circle itself symbolises something which is not 
permitted. If you were to explain what a no smoking symbol looked like 
without saying the circle was red (even if they have no concept of red 
looks like) would give a lesser clue as to its purpose as providing its 
colour. In fact, just saying a red circle, would probably give the idea 
of something that is not permitted, even before saying it has a picture 
of a cigarette in it.


But again, this is only valid if the page in question with the no 
smoking sign is explicitly explaining the appearance and meaning of 
symbols. Otherwise, the purpose of that image is to signal No smoking, 
which would then be a perfectly appropriate ALT text conveying the 
meaning of the image.


P
--
Patrick H. Lauke
__
re·dux (adj.): brought back; returned. used postpositively
[latin : re-, re- + dux, leader; see duke.]
www.splintered.co.uk | www.photographia.co.uk
http://redux.deviantart.com
__
Web Standards Project (WaSP) Accessibility Task Force
http://webstandards.org/
__


**
The discussion list for  http://webstandardsgroup.org/

See http://webstandardsgroup.org/mail/guidelines.cfm
for some hints on posting to the list  getting help
**



Re: [WSG] Web site images question

2006-06-11 Thread Gunlaug Sørtun

David Dixon wrote:

[...]



It would be useful to get a few more opinions on what others believe
 the purpose is...


The purpose of 'text' in the alt-attribute it to expand/complete the
meaning - in context, (as close to) the same as the visible image itself
does.

Describing the image is not what the alt-attribute is there for.
An image may tell a story, and/or strengthen (parts of) the
surrounding content. The alt-text should (ideally speaking) do the same
- or be left out.

Whatever one puts in that alt-attribute should make optimal sense - in
context. If it doesn't, then an alt= is to be preferred in nearly all
cases.

---

The 'graphical' version should not influence or cross-contaminate
the 'text-only' version of a page at the User end.
Few 'text-only' Users can, or do, compare.

The 'text-only' version should carry _all_ that is relevant - and not
much else, while the 'graphical' may provide space for all the relevant,
less-relevant and/or non-relevant visual cues one may want to put in.

---

Images can, and often do, add value - when visible, as visual cues
along with text works well - visually.

Text - followed by an image - followed by more text, may work well in
most cases - for those of us who can see it all and scan in all directions.

---

Providing loss-less alternatives inside alt-attributes is impossible for
most images.
Providing short alternatives in context is somewhat easier, but most
often not necessary, or useful.

Very often the only logical alt-text for an image is a repetition, or
rewrite, of what's already in the text. Such a logical but repetitive
alt-text doesn't add anything of value. Repetitive, rewritten or 'out of
context' text adds noise, so it makes little or no sense to have any
alt-text.

Text - followed by a rewrite of the same or 'an out of context' text
(disguised as an accessibility-improving alt-attribute) - followed by
more (of the same) text, rarely ever make much sense.


Georg
--
http://www.gunlaug.no


**
The discussion list for  http://webstandardsgroup.org/

See http://webstandardsgroup.org/mail/guidelines.cfm
for some hints on posting to the list  getting help
**



Re: [WSG] Web site images question

2006-06-11 Thread John S. Britsios

OK everybody. Now back to reality.

He is a real world testing scenario:

I asked a novice/intermediate(?) Internet user without any known 
disabilities to test a demo page I have created for that purpose (with 
graphics rendered as background in CSS),  and therefore no alt text 
attributes available, and her first question was: What do those symbols 
mean? .


And she continued: John, and why when I go over the graphics, those 
little yellow boxes don't appear?  I saw that often on other pages, were 
they explain what the graphics are about.


To this point, I wanted to add here, that not only experienced users are 
welcome to request any of our services. While to be specific, our tester 
was a paralegal secretary, and she have requested our services a while 
ago, for building a web site for the lawyers company she is working 
with. So I think I cannot consider such employees for example, 
non-targeted group, or?


So, far I still think it would be appropriate to modify the alt 
attributes as in previous mails suggested, but for a long description 
(not longdesc though), I am still not sure about that.


What do you think now?

Thanks again for your kind feedbacks.

Best,

John



David Dixon wrote:
It would be true if this was a page explaining symbols. As it stands, 
though, the purpose of those images is purely an aesthetic 
enhancement to accompany the text. They serve no purpose.


But whatever. It's obvious that until you ascribe meaning to those 
images, we won't agree on this.


You're probably right, unless we agree as the purpose of the image 
(aesthetic or symbolic) then we'll probably just end up arguing our 
cases to verbatim.
It would be useful to get a few more opinions on what others believe 
the purpose is, although in the end, I think John, its down to you to 
decide which camp these images sit in. Not an easy decision based on 
the too and fro that been going on here though :)


Cheers,

David.


**
The discussion list for  http://webstandardsgroup.org/

See http://webstandardsgroup.org/mail/guidelines.cfm
for some hints on posting to the list  getting help
**





--
John S. Britsios
Web Architect  Marketing Consultant

Webnauts Net (Main Office)
Koblenzer Str. 37A
D-33613 Bielefeld

Webnauts Net (U.S. Office)
5 Ivanhoe Drive
Urbana IL 61802

Email: [EMAIL PROTECTED]
Web address: http://www.webnauts.net



**
The discussion list for  http://webstandardsgroup.org/

See http://webstandardsgroup.org/mail/guidelines.cfm
for some hints on posting to the list  getting help
**



Re: [WSG] Web site images question

2006-06-11 Thread Thierry Koblentz
John S. Britsios wrote:
 I asked a novice/intermediate(?) Internet user without any known
 disabilities to test a demo page I have created for that purpose (with
 graphics rendered as background in CSS),  and therefore no alt text
 attributes available, and her first question was: What do those
 symbols mean? .
 
 And she continued: John, and why when I go over the graphics, those
 little yellow boxes don't appear?  I saw that often on other pages,
 were they explain what the graphics are about.

John,
IMO, these questions show that *title* would be more appropriate than alt

---
Regards,
Thierry | www.TJKDesign.com


**
The discussion list for  http://webstandardsgroup.org/

 See http://webstandardsgroup.org/mail/guidelines.cfm
 for some hints on posting to the list  getting help
**



Re: [WSG] Web site images question

2006-06-11 Thread David Dixon
Sorry Georg, I appreciate the time you took in posting your response, 
but I think you misunderstood my question.


I wasn't asking what was the opinion on the appropriateness of alt/title 
text in a general sense, but what the opinion was on the purpose of the 
image that was in question... is it purely aesthetic or does it convey 
its own meaning (i.e. is it symbolic of its own message).


It's purely elementary of course, but the different interpretations are 
a reason why the opinion is split.


Cheers,

David.

Gunlaug Sørtun wrote:

David Dixon wrote:

[...]



It would be useful to get a few more opinions on what others believe
 the purpose is...


The purpose of 'text' in the alt-attribute it to expand/complete the
meaning - in context, (as close to) the same as the visible image itself
does.

Describing the image is not what the alt-attribute is there for.
An image may tell a story, and/or strengthen (parts of) the
surrounding content. The alt-text should (ideally speaking) do the same
- or be left out.

Whatever one puts in that alt-attribute should make optimal sense - in
context. If it doesn't, then an alt= is to be preferred in nearly all
cases.

---

The 'graphical' version should not influence or cross-contaminate
the 'text-only' version of a page at the User end.
Few 'text-only' Users can, or do, compare.

The 'text-only' version should carry _all_ that is relevant - and not
much else, while the 'graphical' may provide space for all the relevant,
less-relevant and/or non-relevant visual cues one may want to put in.

---

Images can, and often do, add value - when visible, as visual cues
along with text works well - visually.

Text - followed by an image - followed by more text, may work well in
most cases - for those of us who can see it all and scan in all directions.

---

Providing loss-less alternatives inside alt-attributes is impossible for
most images.
Providing short alternatives in context is somewhat easier, but most
often not necessary, or useful.

Very often the only logical alt-text for an image is a repetition, or
rewrite, of what's already in the text. Such a logical but repetitive
alt-text doesn't add anything of value. Repetitive, rewritten or 'out of
context' text adds noise, so it makes little or no sense to have any
alt-text.

Text - followed by a rewrite of the same or 'an out of context' text
(disguised as an accessibility-improving alt-attribute) - followed by
more (of the same) text, rarely ever make much sense.


Georg



**
The discussion list for  http://webstandardsgroup.org/

See http://webstandardsgroup.org/mail/guidelines.cfm
for some hints on posting to the list  getting help
**



[WSG] A research on the coding practices in web

2006-06-11 Thread Rene Saarsoo

Hi,

I have conducted a research on roughly a million web pages,
looking for answers to questions like:

* How many pages make use of CSS / JavaScript / ... ?
* How are the main web-technologies used?
  (which html elements, which CSS properties, ...)
* How correct is the code that is used?
  (Valid HTML? Valid CSS2? application/xhtml+xml for XHTML?)

The whole 120-page document about this research is only available
in estonian, but I have put together a shorter version in english,
which mostly consentrates on the results (but is still quite long
read):

http://www.triin.net/2006/06/12/Coding_practices_of_web_pages

As Google did it's Web Authoring Statistics at the beginning
of the year, I don't have much to say about HTML elements and
attributes. And as my JavaScript analyzing program just refused
to work, I don't have much to say about JavaScript either. But
this means the major part of the research deals with CSS -
properties, units, selectors, validation, etc.

Hope you like it,

(any comments are welcome)

Rene Saarsoo


**
The discussion list for  http://webstandardsgroup.org/

See http://webstandardsgroup.org/mail/guidelines.cfm
for some hints on posting to the list  getting help
**



Re: [WSG] Web site images question

2006-06-11 Thread Lachlan Hunt

David Dixon wrote:
If you really are still concerned for size of the alt text though, then 
keep in mind that JAWS appears to have a limit of around 150 characters 
(although im yet to test this),


Joe Clark recommends that alt text be no longer than 1,024 characters 
and I'm quite sure he'd have actually tested it with JAWS and other 
screen readers.

http://joeclark.org/book/sashay/serialization/Chapter06.html#p-905

steer clear of longdesc for its lack of support (plus I feel a longdesc would 
be overkill).


You should see what Joe Clark has to say on the matter.
http://joeclark.org/book/sashay/serialization/Chapter06.html#h2-2415

I recommend everyone (at least those participating in this thread) read 
that entire chapter, as it covers many of the issues being discussed here.


--
Lachlan Hunt
http://lachy.id.au/


**
The discussion list for  http://webstandardsgroup.org/

See http://webstandardsgroup.org/mail/guidelines.cfm
for some hints on posting to the list  getting help
**