Re: Guideline 4 - control rendering

At 11:27 AM 6/23/99 -0400, Ian Jacobs wrote:
>Charles McCathieNevile wrote:
>> 
>> Guideline 4. The first 5 checkpoints cover medium-specific information, and
>> it would seem possible to combine them and the rendering of alternative
>content. 
>
>The topics are: images, background images, video, sound, captions.
>
>Background images: I think this not an alt content issue: it's about
>                   inability to see text (for example) against a
>                   certain background.

This is correct.  Background images reduce the distinctness of visual
content presented against this background.  People who have visual
impairments may need all the distinctness they can get.  They can find the
visual interference introduced by the background image to be a
make-or-break issue.  Most people have enought visual acuity so that a
little such interference is not a problem.

>
>Captions: This is mostly about alt content and could be in guideline 6.
>
>Images, video, sound: I can see wanting to turn them on/off to get
>  them out of the way (reduce distraction) or speed up downloads.
>  But can someone provide a good answer to how they "reduce"
>  accessibility? (The guideline is about that.)

The most obvious gross example is background sounds interfering with the
ability to listen to synthetic speech.  But there is also a problem for the
hearing impaired parallel to the one for background images.

Anything that is distracting can be a make-or-break issue for people with
some cognitive conditions.

Let's take a brief digression into Universal Design.  These problems of
people with disabilities are just the tip of the iceberg.  There is a more
fundamental principle.  The internet is valuable for commerce today because
it extends into the home.  Your home your private space.  Local standards
of decency have been upheld in U.S. pornography law; the same goes for what
constitutes social distance or decorous behavior from home to home.
Standards vary.  The author of the web page does not have unlimited
artistic license to "express herself" when the expression is playing out in
your private space.  The user should have the right to discipline unruly
displays, period.  If the basic setup of the Web as a communication service
is that it has controls for these things, it will be better accepted in
more places.  It is also true that each of these forms of stimulus may be a
make-or-break interference with the user's ability to extract the
information from the document for some class of PWD, and only the
individual user can tell for sure what needs to be turned off.


>> Given that a user will have access to alternative content, (which
>> has to be a P1 requirement), I'm not sure why there is a need to turn off
>> images, although I can understand the need to turn them on.

This is ridiculous.  You will never get to a usable UI if you nickel and
dime the state transitions like this and create Chinese finger-traps all
over the place.

Don't you understand why "open in a new window" on server prompting is so
pernicious?  It means the 'back' command doesn't work, suddenly and without
explanation.  The user's confidence that they are in control is predicated
on the fact that all link-followings are tentative and one can just go back
if it wasn't where you would want to go.  Break that a few times and the
user never knows when they are going to get stuck; and they are suddenly
paranoid about following any link.  The user needs to be able to tinker
with the degrees of freedom that they control. 

>
>How about a situation where you want to get back screen
>real estate (e.g., using screen magnifier) so you turn off images?
>
>>  (4.6, 4.7, and
>> 4.8 deal with particular problems which should probably remain separated
>> out).
>> 
>> Again, I am not sure of the value of being able to urn off the rendering of
>> frames, although I can see value in being able to configure the way in
which
>> they are rendered (It is possible to deal with them as separate windows in
>> some browsers, for example)
>
>I see your point. I think the goal is access to the frame's content
>(covered by another checkpoint). One technique is linearization,
>which may or may not happen when you turn off the frames.

"The way they are rendered" does not cover the topic.  One has to consider
the full state chart, which covers the way they are navigated and
combinations generated as well as how they are laid out.

Likewise, there is the issue of how the user can access the NOFRAMES
section if not by turning off frames support.

Admittedly, emulating a browser that doesn't know from frames could be
viewed as "simply a technique to implement access to the NOFRAMES alternate
content."  On the other hand, since the idea that FRAMES are optional is
enshrined in the DTDs of the Recommendation, and in the behavior
description of NOFRAMES, it does not see inappropriate to make this
technique a requirement.

Al

> _ Ian
>
>-- 
>Ian Jacobs (jacobs@w3.org)   http://www.w3.org/People/Jacobs
>Tel/Fax:                     +1 212 684-1814
>

Received on Thursday, 24 June 1999 08:40:44 UTC