[Accessibility-ia2] new scroll type?

Peter Korn Peter.Korn at Sun.COM
Wed Jun 25 21:29:16 PDT 2008


Hi Pete,

If I understand this correctly, you are suggesting a special role for a 
scrollbar that is at the bottom edge of a window?  Why not just allow 
the AT to obtain that information from the bounding rectangle of the 
scroll bar as compared to it's containing window?


Regards,

Peter Korn
Accessibility Architect,
Sun Microsystems, Inc.
>
> Thanks Alex I think I understand what you are describing, i.e.
>
>     * There is a top level document and the content is large enough
>       such that some of it is above the view port of the application
>       (browser / office app)
>     * There is a nested scrolling control/widget in the top level
>       document and some of that control/widget is also above the
>       viewport of the application.
>     * There is an object in the content of the nested scrolling
>       control/widget and some of that object is above the nested
>       viewport of the nested scrolling control/widget
>
>
> How about a comment like this?
>
>  /** Scroll the bottom edge of the object or substring as close as 
> possible
>   to the bottom edge of the containing scrollable control.  If there is a
>   hierarchy of containing scrollable controls scroll each of the parent
>   scrollable controls in turn starting from the most deeply nested
>   scrollable control until the object or substring is as close as 
> possible
>   to the the bottom edge of the top level window.  
> */
> IA2_SCROLL_TYPE_BOTTOM_EDGE,
>
> *Pete Brunet*
>                                                                          
> IBM Accessibility Architecture and Development
> 11501 Burnet Road, MS 9022E004, Austin, TX 78758
> Voice: (512) 838-4594, Cell: (512) 689-4155
> Ionosphere: WS4G
>
>
>
> *"Alexander Surkov" <surkov.alexander at gmail.com>*
>
> 06/05/2008 01:39 AM
>
> 	
> To
> 	Pete Brunet/Austin/IBM at IBMUS
> cc
> 	Accessibility-ia2 at lists.freestandards.org
> Subject
> 	Re: [Accessibility-ia2] new scroll type?
>
>
>
> 	
>
>
>
>
>
> Hi, Pete.
>
> Sorry I put two examples but gave one picture. But you talk correct 
> things and they are coming with used firefox algorithm.
>
> About window and application client area I meant they are both good if 
> you use them in the "screen" meaning. But when I see them I think 
> about top level view (the view for browser or view for documents in 
> the case of DOM window term). Note in this sense, application client 
> area or window may be not scrollable but the same time the scrolled 
> object may be contained by scrolled area.
>
> ±±±±±±±±±±±±±±±±±±±±±±
> | scrolled object
> |!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
> | nested scrollable area
> |
> |______________________________
> | application client scrollable area
> |
> |±±±±±±±±±±±±±±±±±±±±±±±±±
> |
> |!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
> |
> |________________________________
>
> Here we should scroll top of "scrolled object" to top edge of "nested 
> scrollable area" and top of "nested scrollable area" to top edge of 
> "application client scrollable area". Possibly we should mention we 
> have in view all nested scrollable areas that contain our scrolled 
> object.
>
> Alex.
>
>
>
> On Thu, Jun 5, 2008 at 1:08 PM, Pete Brunet <_brunet at us.ibm.com_ 
> <mailto:brunet at us.ibm.com>> wrote:
>
> Alexander,
>
> > ...then scroll_type_anywhere is equivalent to scroll_type_bottom_right
> I don't understand this.  I think, but I might be wrong, that you are 
> describing these two examples.  In the first one the left edges are 
> aligned.  In the second one the tops are aligned.
> ______________________
> | scrolled object
> |________________
> | visible area
> |
>
> ______________________
> | scrolled object   | visible area
> |                                 |
> |                                 |
>
>
> In the first case you'd just move the object straight down.  In the 
> second case you'd move the object to the right.  It would be like a 
> top_edge and left_edge request, not a bottom_right request.  But I 
> probably don't understand what you are thinking.
>
> Regarding "window" vs "application's client area", I didn't understand 
> the original issue, i.e. you said, "the word window is reserved by 
> DOM".  Please explain.  I assumed that there was some reason why I 
> couldn't use the word window so I used something equivalent.
>
> >Would it be better to use 'scrollable area'
>
> To me, "scrollable area" is the same as "window" or "application's 
> client area".  
>
> >emphasize there may be more than one view in hierarchy that should be 
> scrolled to make the object visible?
>
> Could you give me an example so I can understand this?
>
> Thanks,
> *Pete Brunet*
>                                                                        
> IBM Accessibility Architecture and Development
> 11501 Burnet Road, MS 9022E004, Austin, TX 78758
> Voice: (512) 838-4594, Cell: (512) 689-4155
> Ionosphere: WS4G
>
>
> *"Alexander Surkov" <**_surkov.alexander at gmail.com_* 
> <mailto:surkov.alexander at gmail.com>*>*
>
> 06/04/2008 09:46 PM
>
> 	
> To
> 	Pete Brunet/Austin/IBM at IBMUS
> cc
> 	_Accessibility-ia2 at lists.freestandards.org_ 
> <mailto:Accessibility-ia2 at lists.freestandards.org>
> Subject
> 	Re: [Accessibility-ia2] new scroll type?
>
>
>
>
> 	
>
>
>
>
>
>
> Hi, Pete.
>
> That sounds better. I think minimal amount is essential. And in the 
> case when scrolled object is above of visible area like
>
> ______________________
> | scrolled object
> |      ________________
> |      | visible area
> |      |
>  
> scroll_type_anywhere is equivalent to scroll_type_top_left. But if 
> top/left edge of the scrolled object is visible then 
> scroll_type_anywhere is equivalent to scroll_type_bottom_right if 
> top/left edge stay visible.
>
> Sorry but I didn't catch a difference between old comment and reword 
> comment for scroll_type_bottom_edge excepting it started to be more 
> complicated :).
>
> I think application's client area sounds like previously used the word 
> 'window'. Would it be better to use 'scrollable area' or something to 
> emphasize there may be more than one view in hierarchy that should be 
> scrolled to make the object visible?
>
> Alex.
>
>
> On Wed, Jun 4, 2008 at 12:57 AM, Pete Brunet <_brunet at us.ibm.com_ 
> <mailto:brunet at us.ibm.com>> wrote:
>
> Alex, I changed it a bit.  How about this?
>
> /** Scroll the object or substring a minimal amount such that as much 
> as possible
>  of the object or substring is within the application's client area.
> */
>
> Is it important that the object only be moved a minimal amount?  For 
> example, if the object was off screen to the top and to the left is it 
> important that the action be equivalent to  IA2_SCROLL_TYPE_TOP_LEFT? 
>  Or would it be OK for the object to be placed anywhere on the screen. 
>  I suspect that minimal movement *is* important, i.e. it's likely that 
> as much of the current screen content be preserved while string 
> bringing the object into view.
>
> This brings up an interesting question.  The list of constants is:
>
> IA2_SCROLL_TYPE_TOP_LEFT,                
> IA2_SCROLL_TYPE_BOTTOM_RIGHT,
> IA2_SCROLL_TYPE_TOP_EDGE,
> IA2_SCROLL_TYPE_BOTTOM_EDGE,
> IA2_SCROLL_TYPE_LEFT_EDGE,
> IA2_SCROLL_TYPE_RIGHT_EDGE,
> IA2_SCROLL_TYPE_ANYWHERE
>
> If I use bottom right and the object is above the client area then it 
> would be valid with respect to the spec but not acceptable to the user 
> to move the object such that only the bottom pixel or row of pixels is 
> viewable.  Do ATs know to not use that constant when the object is 
> above the view port?  Ditto with bottom left.
>
> Also, I reworded the 4 edge constants to say that the full edge must 
> be made visible.  For example,
>
> /** Scroll as much as possible of the bottom edge of the object or 
> substring
>  as close as possible to the bottom edge of the application's client 
> area.
> */
> IA2_SCROLL_TYPE_BOTTOM_EDGE,
>
> Otherwise the app could just move a single pixel or row of pixels on 
> screen, i.e. think of the case when the object is off screen to the 
> bottom and to the left. *
>
> Pete Brunet*
>                                                                        
> IBM Accessibility Architecture and Development
> 11501 Burnet Road, MS 9022E004, Austin, TX 78758
> Voice: (512) 838-4594, Cell: (512) 689-4155
> Ionosphere: WS4G
>
> *"Alexander Surkov" <**_surkov.alexander at gmail.com_* 
> <mailto:surkov.alexander at gmail.com>*>*
>
> 06/02/2008 09:59 PM
>
> 	
> To
> 	Pete Brunet/Austin/IBM at IBMUS
> cc
> 	_Accessibility-ia2 at lists.freestandards.org_ 
> <mailto:Accessibility-ia2 at lists.freestandards.org>
> Subject
> 	Re: [Accessibility-ia2] new scroll type?
>
>
>
>
>
> 	
>
>
>
>
>
>
>
> Hi, Pete.
> I think the word window is reserved by DOM, I would suggest to use 
> here the word "screen" or "view area" instead. The phrase "can be 
> placed anywhere" makes me scary a bit because it sounds undefined. 
> Would it be better to use "scroll minimum amount to make it visible" 
> or something like this?
>
> Alex.
>
> On Tue, Jun 3, 2008 at 5:11 AM, Pete Brunet <_brunet at us.ibm.com_ 
> <mailto:brunet at us.ibm.com>> wrote:
>
> Please are the following constant and description which I added this 
> to the spec:
>
> /** Scroll the object or substring such that it is placed fully within 
> the
> window (or as much as possible).  The object or substring can be placed
> anywhere within the window.
> */
> IA2_SCROLL_TYPE_ANYWHERE *
>
> Pete Brunet*
>                                                                      
> IBM Accessibility Architecture and Development
> 11501 Burnet Road, MS 9022E004, Austin, TX 78758
> Voice: (512) 838-4594, Cell: (512) 689-4155
> Ionosphere: WS4G
>
> *"Alexander Surkov" <**_surkov.alexander at gmail.com_* 
> <mailto:surkov.alexander at gmail.com>*>*
>
> 05/27/2008 08:36 PM
>
> 	
> To
> 	Pete Brunet/Austin/IBM at IBMUS
> cc
> 	_Accessibility-ia2 at lists.freestandards.org_ 
> <mailto:Accessibility-ia2 at lists.freestandards.org>
> Subject
> 	Re: [Accessibility-ia2] new scroll type?
>
>
>
>
>
>
> 	
>
>
>
>
>
>
>
>
> I have looked into mozilla code and there our constant is much similar 
> with IA2_SCROLL_TYPE_BUBBLE_ONLY_IF_NECESSARY. We could implement 
> another constant as long as we know its meaning and for sure if it's 
> necessary :) Because SCROLL_TYPE_ANYWHERE means AT doesn't care where 
> accessible object will be (just in view). But I haven't idea how can 
> IA2_SCROLL_TYPE_BUBBLE be used.
>
> Alexander.
>
>
> On Tue, May 27, 2008 at 11:55 PM, Pete Brunet <_brunet at us.ibm.com_ 
> <mailto:brunet at us.ibm.com>> wrote:
>
> Alexander, Would SCROLL_TYPE_ANYWHERE be equivalent to David Bolter's 
> IA2_SCROLL_TYPE_BUBBLE_ONLY_IF_NECESSARY?  
>
> And do you see David's IA2_SCROLL_TYPE_BUBBLE as something you could 
> implement? *
>
> Pete Brunet*
>                                                                      
> IBM Accessibility Architecture and Development
> 11501 Burnet Road, MS 9022E004, Austin, TX 78758
> Voice: (512) 838-4594, Cell: (512) 689-4155
> Ionosphere: WS4G
>
> *"Alexander Surkov" <**_surkov.alexander at gmail.com_* 
> <mailto:surkov.alexander at gmail.com>*>*
>
> 05/25/2008 08:54 PM
>
> 	
> To
> 	Pete Brunet/Austin/IBM at IBMUS
> cc
> 	_Accessibility-ia2 at lists.freestandards.org_ 
> <mailto:Accessibility-ia2 at lists.freestandards.org>, 
> _Accessibility at lists.freestandards.org_ 
> <mailto:Accessibility at lists.freestandards.org>
> Subject
> 	Re: [Accessibility-ia2] new scroll type?
>
>
>
>
>
>
>
> 	
>
>
>
>
>
>
>
>
>
> In Gecko accessibility API we have constant named SCROLL_TYPE_ANYWHERE 
> which allows to scroll accessible into view. We can map it into new 
> IA2 constant.
>
> Alexander.
>
>
> On Sat, May 24, 2008 at 8:13 AM, Pete Brunet <_brunet at us.ibm.com_ 
> <mailto:brunet at us.ibm.com>> wrote:
>
> For IA2::scrollTo it would be nice to have an enum value for 
> IA2ScrollType that was not limited to an edge or corner in the case 
> that the AT doesn't care how it's brought on screen. Does anyone have 
> some insight into if a new constant  is needed? *
>
> Pete Brunet*
>                                                                    
> IBM Accessibility Architecture and Development
> 11501 Burnet Road, MS 9022E004, Austin, TX 78758
> Voice: (512) 838-4594, Cell: (512) 689-4155
> Ionosphere: WS4G
>
> _______________________________________________
> Accessibility-ia2 mailing list_
> __Accessibility-ia2 at lists.linux-foundation.org_ 
> <mailto:Accessibility-ia2 at lists.linux-foundation.org>_
> __https://lists.linux-foundation.org/mailman/listinfo/accessibility-ia2_
>
>
>
> _______________________________________________
> Accessibility-ia2 mailing list_
> __Accessibility-ia2 at lists.linux-foundation.org_ 
> <mailto:Accessibility-ia2 at lists.linux-foundation.org>_
> __https://lists.linux-foundation.org/mailman/listinfo/accessibility-ia2_
>
>
> _______________________________________________
> Accessibility-ia2 mailing list_
> __Accessibility-ia2 at lists.linux-foundation.org_ 
> <mailto:Accessibility-ia2 at lists.linux-foundation.org>_
> __https://lists.linux-foundation.org/mailman/listinfo/accessibility-ia2_
>
>
> _______________________________________________
> Accessibility-ia2 mailing list_
> __Accessibility-ia2 at lists.linux-foundation.org_ 
> <mailto:Accessibility-ia2 at lists.linux-foundation.org>_
> __https://lists.linux-foundation.org/mailman/listinfo/accessibility-ia2_
>
>
> _______________________________________________
> Accessibility-ia2 mailing list_
> __Accessibility-ia2 at lists.linux-foundation.org_ 
> <mailto:Accessibility-ia2 at lists.linux-foundation.org>_
> __https://lists.linux-foundation.org/mailman/listinfo/accessibility-ia2_
>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Accessibility-ia2 mailing list
> Accessibility-ia2 at lists.linux-foundation.org
> https://lists.linux-foundation.org/mailman/listinfo/accessibility-ia2
>   

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.linux-foundation.org/pipermail/accessibility-ia2/attachments/20080625/5b2d8ff8/attachment-0001.htm 


More information about the Accessibility-ia2 mailing list