W3C home > Mailing lists > Public > w3c-wai-ua@w3.org > October to December 1999

Re: direct and spatial mapping to functionalities

From: Charles McCathieNevile <charles@w3.org>
Date: Sun, 3 Oct 1999 22:52:27 -0400 (EDT)
To: thatch@us.ibm.com
cc: Marja-Riitta Koivunen <marja@w3.org>, w3c-wai-ua@w3.org
Message-ID: <Pine.LNX.4.10.9910032208590.28671-100000@tux.w3.org>
This is true, if done correctly. However I think there are some issues to be
resolved. 

  1. As Marja has pointed out, implementing mousekeys would satisfy the
  checkpoint with respect to IE/Opera/Netscape (and a lot of other software)
  without satisfactorily solving the problem of access.

  2. A minimum requirement is not just to be able to use the keyboard, but to
  provide access to the functionalities through the standard keyboard API.
  Although this is the obvious way to do it, it should be stateed explicitly
  in the text of the requirement.

I think the real question is the interface model - creating a model which
relies in a fundamental way on the user dealing with spatial relationships
rather than allowing serial or "by-name" interaction with the interface
components is the real problem, I think.

One of the problems is the way that HTML events were originally implemented,
using a fundamentally mouse-based model of the user interface. It should
probably be a requirement for HTML-consuming browsers that they work with an
improved device-specific event model as soon as one is available. (There is
work on this going on in the DOM working group, among others).

Please do not stop Marja. I think you have identified a problem we didn't
see, and I think you have pointed to the answer too.

Charles McCN


On Sun, 3 Oct 1999 thatch@us.ibm.com wrote:

  
  
  MK: Does everybody else understand this perfectly and agree that the current
  wording reflects it? Then I will stop.
  
  JT: I am not going to ask you to stop, but even if 2.1 is flawed
  (and I don't think it is) it should remain as is. The wording is
  unequivocal and clear. It is, in my opinion, the most important
  checkpoint for access to software for multiple disabilities. Water
  it down or generalize it and it will loose its focus and force.
  
  2.1 By default and without additional customization, ensure that all
  functionalities offered by the user agent are accessible using the keyboard.
  
  Jim Thatcher
  IBM Special Needs Systems
  www.ibm.com/sns
  HPR Documentation page: http://www.austin.ibm.com/sns/hprdoc.html
  thatch@us.ibm.com
  (512)838-0432
  
  
  Marja-Riitta Koivunen <marja@w3.org> on 09/29/99 10:56:43 AM
  
  To:   Ian Jacobs <ij@w3.org>
  cc:   ian@w3.org, w3c-wai-ua@w3.org
  Subject:  Re: direct and spatial mapping to functionalities
  
  
  
  
  At 11:33 AM 9/29/99 -0400, Ian Jacobs wrote:
  >Marja-Riitta Koivunen wrote:
  >>
  >> While thinking about conformance I was looking the guidelines and
  >> checkpoints again. I still don't like the word keyboard in guideline 2. I
  >> also think keyboard access is not what we want to say in  many checkpoints
  >> e.g. in
  >>
  >> 2.1 By default and without additional customization, ensure that all
  >> functionalities offered by the user agent are accessible using the
  keyboard.
  >>
  >> So you could use the keyboard arrow keys to point and some other key to
  >> select and still conform? Or what about my laptop keyboard with a finger
  >> mouse built into it?
  >>
  >> I think we want to say something about offering direct mapping from input
  >> device keys to the functionalities as opposite to spatial mapping with
  >> pointing and graphical objects. In the first case we usually have many keys
  >> or key combinations that the user needs to remember but no need to point or
  >> see. In the latter case we need to remember just few keys and some way to
  >> point in 2D (or 3D). If we can present the activating of functionalities
  >> with graphical objects or by using force feedback it often helps memory but
  >> it is slower to get to the functions.
  >>
  >> I think both mappings are important. The point&click UI with explorable
  >> memory aid (e.g. graphical  objects, sound map, force feedback map) helps
  >> cognitively disabled (and everyone with human memory) the direct mapping
  >> helps motorically disabled because some key or morse code etc. can be
  >> mapped directly to the function without need to go through the spatially
  >> located object.
  >
  >There is an analogy with serial access to links (which provides
  >context as-you-go) and direct access (which is faster, but requires
  >more experience). Explaining the utility of both for access
  >to UA functionality would be useful (we already do so for navigation).
  >
  >However, I think your abstraction overlooks the need that motivated
  >this Guideline: assistive technology today relies on software using
  >the standard OS keyboard API (please correct me if I'm wrong). This
  >Guideline is less abstract than others since it addresses today's
  >technology and today's requirements. Device-independence captures
  >the principle while talking about the keyboard API captures today's
  >need.
  
  Guideline is not a checkpoint, so it is not what people check, it does not
  even have priorities. It is put higher up because it is important, still
  this guideline does not need to be conformed always. So I'm confuced.
  
  Then I look the actual checkspoint under the guideline, such as
  
  2.1 By default and without additional customization, ensure that all
  > functionalities offered by the user agent are accessible using the keyboard.
  >
  > So you could use the keyboard arrow keys to point and some other key to
  > select and still conform? Or what about my laptop keyboard with a finger
  > mouse built into it?
  
  And I'm even more confused. Here I agree we should mention keyboard, but
  when we have mentioned it, it seems not to help. Can I use some of my
  keyboard keys (or even my keyboard mouse for pointing) and conform to this?
  I don't think this is what we want.
  
  Does everybody else understand this perfectly and agree that the current
  wording reflects it? Then I will stop.
  
  Marja
  
  
  >I am not yet convinced (but still open!) that presenting the
  >Guideline as requiring direct v. serial access to UA functionality
  >will adequately address the requirement of today's technology.
  >
  >Please let me know if my comments reflect an understanding of
  >your suggestion.
  >
  >Thank you,
  >
  > - Ian
  >
  >
  >> A separate thing is then how to present all this. If the user can see she
  >> can have memory aid on the screen (or even paper) also for directly mapped
  >> keyboard events, if she cannot she needs to rely more on memory. On the
  >> other hand she may use spatial mapping and exhaustive spatial search with
  >> sound or force feedback to help her memory. The graphical object model
  >> provides the memory aid naturally but can also be badly designed.
  >>
  >> Marja
  >
  >--
  >Ian Jacobs (jacobs@w3.org)   http://www.w3.org/People/Jacobs
  >Tel/Fax:                     +1 212 684-1814
  >Cell:                        +1 917 450-8783
  >
  
  
  
  

--Charles McCathieNevile            mailto:charles@w3.org
phone: +1 617 258 0992   http://www.w3.org/People/Charles
W3C Web Accessibility Initiative    http://www.w3.org/WAI
MIT/LCS  -  545 Technology sq., Cambridge MA, 02139,  USA
Received on Sunday, 3 October 1999 22:52:28 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 October 2009 06:49:34 GMT