W3C home > Mailing lists > Public > wai-xtech@w3.org > April 2008

RE: Style guide: tristatecheckbox

From: Sailesh Panchang <sailesh.panchang@deque.com>
Date: Mon, 14 Apr 2008 14:17:18 -0400
To: "'Aaron M Leventhal'" <aleventh@us.ibm.com>
Cc: <wai-xtech@w3.org>, <wai-xtech-request@w3.org>
Message-ID: <48039d88.43e2220a.7639.fffffe6a@mx.google.com>
In reality a 3-state checkbox is  a poorly designed control.

In the example  with a tree control and checkbox / radio buttons for
underlying components, the installation process might require the user to
indicate selections against all components (mandatory). 

Or  it may not be mandatory in which case default selections would be used I
assume.

 

If it is mandatory, it does not matter if the user has not indicated
selections against 1 or all components. Till all are indicated, the state of
the parent control is incomplete or unchecked. A more useful state for a
parent control of this nature would be one which indicates count (out of a
max possible) that are answered like 7/12 i.e. 7 have been answered and 5
are unanswered.

 

Radio and checkbox controls are essentially boolean and should be  retained
that way.

Microsoft regards       a tri-state checkbox is a group of three radio
buttons, two of which represent the false and the true states and a third
which stands for the undefined state. Not sure why  this is needed. A null
value is returned for the undefined state anyway.

Not sure if this is open for debate here.

 

Sailesh

  _____  

From: Aaron M Leventhal [mailto:aleventh@us.ibm.com] 
Sent: Monday, April 14, 2008 10:20 AM
To: sailesh.panchang@deque.com
Cc: wai-xtech@w3.org; wai-xtech-request@w3.org
Subject: RE: Style guide: tristatecheckbox

 


Salesh, 

It's not common, but here is a use case on the desktop today: 
In an installer, you may have a tree view of components to select. Each item
in the tree view is a tristate checkbox. If only some of the some items are
checked, then the checkbox is shown as "mixed". 

- Aaron 






"Sailesh Panchang" <sailesh.panchang@deque.com> 
Sent by: wai-xtech-request@w3.org 

04/14/2008 04:24 PM 


Please respond to
<sailesh.panchang@deque.com>


To

Aaron M Leventhal/Cambridge/IBM@IBMUS, <wai-xtech@w3.org> 


cc

 


Subject

RE: Style guide: tristatecheckbox

 


 

 




I question the concept of a partially checked checkbox. 
Checkbox and radio buttons are meant to hold  just the 2 boolean values: yes
or no. 
If one can have  a 3 state checkbox  (to correspond to Yes, No, Not sure),
then why not a 5 state checkbox (that corresponds with something like
Strongly agree, Agree, Not decided, Disagree, Strongly disagree)? 
  
  

Sailesh Panchang
Accessibility Services Manager (Web and Software)
Deque Systems Inc. (www.deque.com)
11130 Sunrise Valley Drive, Suite #140,
Reston VA 20191
Phone: 703-225-0380 (ext 105)
E-mail: sailesh.panchang@deque.com 

 

  _____  


From: wai-xtech-request@w3.org [mailto:wai-xtech-request@w3.org] On Behalf
Of Aaron M Leventhal
Sent: Friday, April 11, 2008 6:11 AM
To: wai-xtech@w3.org
Subject: Style guide: tristatecheckbox 
  

The below definition conflicts with itself. 
If the item is partially checked, space rotates through checked, unchecked
and then partially checked again. 
However, the first bullet says if it's not checked that space  checks it! 

Is the implementation supposed to have 2 different code paths for unchecked
checkboxes depending on whether it was originally partially checked? 

- Aaron 

*	Three State Check Box 

*	If not checked, space checks the check box 
*	If checked, space unchecks the check box 
*	If partially checked, space will rotate through checked, unchecked,
and partially checked states.

  
Received on Monday, 14 April 2008 18:09:00 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:51:35 UTC