W3C home > Mailing lists > Public > w3c-wai-ua@w3.org > April to June 2000

Re: Problems with proposed checkpoint on synchronized views

From: Jon Gunderson <jongund@ux1.cso.uiuc.edu>
Date: Mon, 08 May 2000 15:27:36 -0500
Message-Id: <4.3.1.2.20000508151629.00c2c430@staff.uiuc.edu>
To: Ian Jacobs <ij@w3.org>, w3c-wai-ua@w3.org
Responses in JRG:
At 03:09 PM 5/6/00 -0400, Ian Jacobs wrote:
>Hello,
>
>At the 2 May teleconference we discussed a new requirement for
>synchronized views. In trying to add a checkpoint
>to the document, I have come across the following questions:
>
>1) What does it mean to synchronize a view? Is it any more than
>    saying that the current focus in one viewport should follow
>    the current focus in the other (as well as possible)? Are there
>    other aspects to synchronization? I suspect not, for if the views
>    must be exact duplicates always, I don't see the point of having
>them.
>
>    Summary: Synchronized views means synchronized content focuses.


JRG: The main context of the issue of synchronization came up in was in 
source or dom views. The need was to have a way to orient the user to the 
source or dom view of what had generated the selected content in their 
primary view.


>2) Which views should be synchronized? I believe that the user agent
>    should <strong>not</strong> synchronize all views that happen to
>    refer to the same content. (In fact, it's even difficult to
>    say what we mean by "the same content".
>    What if two different URIs designate copies
>    of the same document? The user agent should not be required to
>    know that they are copies and that two views are rendering the
>    same content from different resources). Even when I open two views
>    for the same URI, I do not want the focuses to follow each other
>    automatically. In fact, usually I don't want them to follow each
>    other since I opened two different views to navigate the same
>    content in two different ways.
>
>    Furthermore, synchronized focus changes may be disorienting
>    if the user is unaware that associated focuses are meant to
>    change together.
>
>    Summary: Two views should automatically be synchronized simply
>             because they display the same resource.

JRG: Probably the minimum requirement would be synchronizing a source or 
dom view with the primary view.   I

>3) When does synchronization stop? Suppose I navigate to a form
>    control and submit a form and the focus is following my
>    actions in another view that also contains the form. Should the
>    form be submitted in the second view? I don't think in the general
>    case it should. What about following a link in one view - should
>    the link be followed in the other view? What if the link isn't
>    even visible (i.e., it's hidden since the renderings are
>    different)? When should the focuses cease to follow each other?

JRG: I am not sure that two primary views of the same resources should be 
synchronized.  I think if we want to do this the synchronized view needs to 
be some type of view-only view or something the user doesn't interact with 
directly.  The main purpose of the synchronized view was to allow access to 
"machine" content in a way that was easy for the user.  In our discussions 
these were view only views.  The user did not interact with them.

>4) When do we want to require synchronization? We have heard some
>    good examples of using synchronized views, such as synchronizing
>    a link view with a primary view. In this case, the user should
>    be able to navigate the links and would like the focus to follow
>    in the primary view. That seems useful to me, but that happens
>    to be a useful feature provided by the user agent. How many
>    synchronized views are we requiring? One? More than one? Which
>    ones? As pointed out in point 2, I don't believe the answer
>    should be "synchronize all views". If we only require one
>    synchronized view, what should it contain? The links view
>    mentioned above may be implemented through style sheets, and
>    synchronized views in which users can adjust rendering is indeed
>    a powerful tool. However, user agents aren't required to
>    implement a links view with style sheets. Are we going to
>    require any other views than the outline view?
>
>    Summary: Other than the outline view, we don't require
>             the user agent to offer any other views.
>             Therefore, I don't see how we can require
>             synchronization between views (other than the
>             primary and outline views) when we don't
>             know what the semantics of those views will be.


JRG: I think the outline view right now would be a good place to require 
the synchronization.



>We have decided to require an outline view but have not
>required that view to be navigable (though we have suggested
>a navigable outline view would be a good thing). Do we wish
>to now require that the outline view be navigable and that
>the focuses should be synchronized?
>
>In short, I think that we need to reconsider the resolution to
>add a checkpoint requiring the user agent to synchronize views.
>While synchronized views are certain powerful techniques for
>navigation, I don't believe we have considered the problem
>fully enough to add a checkpoint this late in the process. I propose
>instead that we explain view synchronization in the techniques
>document and provide the links view example.
>
>  - Ian


JRG: I think it would be wise for the group to drop the synchronization 
checkpoint and consider synchronization as part of the minimum requirement 
for 7.6 and/or 8.5.

Jon Gunderson, Ph.D., ATP
Coordinator of Assistive Communication and Information Technology
Chair, W3C WAI User Agent Working Group
Division of Rehabilitation - Education Services
College of Applied Life Studies
University of Illinois at Urbana/Champaign
1207 S. Oak Street, Champaign, IL  61820

Voice: (217) 244-5870
Fax: (217) 333-0248

E-mail: jongund@uiuc.edu

WWW: http://www.staff.uiuc.edu/~jongund
WWW: http://www.w3.org/wai/ua
Received on Monday, 8 May 2000 16:27:39 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 October 2009 06:50:03 GMT