W3C home > Mailing lists > Public > w3c-wai-gl@w3.org > January to March 2001

RE: javascript and URLs (new window)

From: Al Gilman <asgilman@iamdigex.net>
Date: Sun, 04 Mar 2001 10:46:02 -0500
Message-Id: <200103041526.KAA4948504@smtp2.mail.iamworld.net>
To: <w3c-wai-gl@w3.org>
Cc: wai-tech-comments@w3.org

On the one hand, there are clear attractions to making window spawning easy in
the environment of a user who has full use of a many-pixel display.  On the
other hand, there are downsides to too-easy window proliferation (or too
silent) for those not using such a device in their interaction environment. 
This is a corollary of the fact that in the absense of the screen as a
persistent shared resource between computer and user, the navigation
options or
shape of the navigable space has to be something that the user can produce
from
their recall memory.

See

 HCI Fundamentals and PWD Failure Modes
 <http://trace.wisc.edu/docs/ud4grid/#_Toc495220368>http://trace.wisc.edu/d
ocs/ud4grid/#_Toc495220368

for a summary of the related theory.  

The approach that is reflected in the User Agent guidelines at the moment is a
little like styling for actions.  Opening and moving to a new window is not
forbidden, but the process by which it happens can be cast by the user into
alternative patterns.  Sometimes the system will JustDoIt for the script
author.  But if the user wants a more controlled process, a notification or
confirmation request is required before the system will take this action.

Details of this functionality in the UAAG are still being hammered out in the
discussion of issue resolutions.  However, the general approach described
above
is clearly present in the current state of the document.  See for example the
topic discussed first on 2 March from last week's meetings.  This is directly
related.  The performance issue is how the user can trust the user
interface to
be stable, aside from the effects of actions that the user consciously
performs
to adjust or configure that user interface.

 Minutes from 1-2 March 2001 UAWG Face-to-face meeting
 <http://www.w3.org/WAI/UA/2001/03/ua-minutes>http://www.w3.org/WAI/UA/2001
/03/ua-minutes

But in general, review the Guideline 4 "checkpoints for user interface
accessibility"

 4. Ensure user control of rendering and viewport
 <http://www.w3.org/WAI/UA/WD-UAAG10-20010224/#gl-user-control-styles>http:
//www.w3.org/WAI/UA/WD-UAAG10-20010224/#gl-user-control-styles

The basic requirement is that the user be able to tune the interaction
space so
that if an application wants to spawn an independent thread of activity, it
has
to "may I" the user. This is because under some circumstances the user
themself
is the only agent that should be able to  authorize the creation of such an
object.  The user needs to be able to keep the set of separate things
happening
that can raise events or otherwise require the user's attention to a short
list
that they can mentally manage.  In GUI systems, one can usually skip this
check.  The application can simply register the new window with the window
manager.  However, in other interaction environments, such as voice dialog,
interactive voice response, or a graphical user interface session managed
through a screen reader, it can be better to have the user themselves as the
gatekeeper for the creation of new sub-sessions, or 'rooms' in the interaction
space.

Note that to the extent that one can force the opening of a new window, and
open a new window without standard controls, by means of some of the attribute
values in the frame language in HTML, there is an XHTML 2.0 frames-reform
issue
embedded here.  As well as the obvious large "civilizing the script interface"
issue.

Al

At 08:10 AM 2001-03-04 -0500, Josh Krieger wrote:
>I couldn't disagree more.  This is one of those "until user agents..."
>guidelines.  There are many good reasons to use pop-up windows for sighted
>users that make sites substantially easier to use and more accessible -- why
>do you think so many people use them.  The fact that pop-ups on the screen
>distract a screen-reader is not so helpful in content guidelines, it's a UA
>issue and could easily be handled by a good piece of software (anyone know
>what the IBM reader does here?).  It would be good to hear from some blind
>folks out there how much of an issue this really is for them.
>
>+ Josh +
>
>-----Original Message-----
>From: w3c-wai-gl-request@w3.org
[<mailto:w3c-wai-gl-request@w3.org%5DOn>mailto:w3c-wai-gl-request@w3.org]On
>Behalf Of Lisa Seeman
>Sent: Sunday, March 04, 2001 1:39 AM
>To: Robert Neff; wai-gl
>Subject: Re: javascript and URLs
>
>
>I think that there is consensus that opening a new window is not, ever, a
>good thing. It confuses navigation. (Imagine a non sited person who is moved
>without her knowledge to a pop up window and then tries the back button -
>disorienting the user.)
>Lisa
>-----Original Message-----
>From: Robert Neff <robneff@home.com>
>To: wai-gl <w3c-wai-gl@w3.org>
>Date: Saturday, March 03, 2001 9:45 PM
>Subject: javascript and URLs
>
>
>>wendy, have we considered javascript opening URLs in new new windows (that
>>are not pop-ups) as a techniques or other?  was just thinking about the
>>number of web sites includeing news sites that do this.  you click on the
>>URL link and it opens another browser window.  what if javascript is turned
>>off? or do we need no script here?
>>
>>comments?
>>
>>
>  
Received on Sunday, 4 March 2001 10:27:11 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:47:09 GMT