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

Comments on keyboard mapping among assistive technology tools

From: Rob Relyea <rrelyea@MICROSOFT.com>
Date: Wed, 7 Apr 1999 09:13:11 -0700
Message-ID: <3FF8121C9B6DD111812100805F31FC0D0D516EE3@RED-MSG-59>
To: "'Harvey Bingham'" <hbingham@ACM.org>, w3c-wai-ua@w3.org
Here is a summary of my comments to Harvey posting of 3/26 about Keyboard
mapping among assitive technology tools:
RR: Consistency with operating system conventions and major applications on
the same operating system is good for users and therefore is desirable for
application verndors.  AT software and application software should work to
provide consistency that is good for users.
Setting a standard for all platforms does not seem realistic due to
different keyboard accelerators, etc...

RR: This requirement sounds like it would be in the AT vendors interest to
do as well as the users.  Reducing the cost and confusion of switching from
your competitors product to yours will help your business.  Example: Word
provided keyboard shortcuts that emulated WordPerfect to help ease the
transition for users.

RR: Agreed.  Providing keyboard remapping facilities for User Agents is a
desirable feature.

RR: This sounds like something the AT vendor should do.  User Agents could
provide a method for different profiles to be used.

-----Original Message-----
From: Harvey Bingham [mailto:hbingham@ACM.org]
Sent: Friday, March 26, 1999 8:12 AM
To: w3c-wai-ua@w3.org
Subject: Keyboard mapping among assistive technology tools

1. Observations:

1.1 There is little consistency in keyboard mappings for special commands 
and functions across the different assistive technology (AT) software,
visual user agents, and operating systems. 

1.2 Alan Cantor notes some 600 combinations of special keys and regular 
keys are available. A typical application only uses a few of these, to
perform application-specific functions, possibly augmented by user 
assignment. The effect of these may depend on the point of regard, the 
focus, the current selection, the window or toolbar, application, 
operating system, etc. 

1.3 A user exploring a new AT package may first need to absorb the keyboard
mappings. A user familiar with one AT package will have less initial
learning if the familiar keyboard could be used, at least as a starting
set. Then only the differences would need to be absorbed.

2. Possible solution:

WAI, as an application-, vendor-, and system-neutral organization, may
be in a position to recommend a common core of keyboard functionality
for assistive technology use.

2.1 Each supplier should permit keyboard customization, and in fact 
provide starter customizations for persons familiar with other packages.

Benefit: Reduced startup effort to transition from a prior familiar AT 

2.2 A new AT supplier would presumably consider the keyboard assignment 
work of others, recognize the desirability of commonality, and explain 
any differences.
Benefit: Keep what works, augment only for what's new.

2.3 A bridge to using more than one package would be to allow a user
personality profile that would allow identification of mappings from 
function to its familiar keyboard mapping. Persons with limited 
finger choices (such as one-handed) will need keyboard remapping, so
it may already be included. 

Benefit: Flexibility of user choice.

3. Problems across user agents, operating systems, and assistive 

Users comfortable with one keyboard mapping will be unhappy making any
changes. [Most English typists still use a QWERTY keyboard, in spite of the
demonstrated enhancements to typing rate and accuracy using the DVORAK
keyboard! In DOS/WINDOWS the CTRL+ALT+DEL remains an essential part of
daily life.]

3.1 Slight differences in effect may exist among implementations of 
similarly named concepts and functions.

3.2 Overlapping key assignments from different user agents for different 
keyboard functions, possibly with disastrous consequences if confused, or
mapped in ignorance of the differences. 

3.3 Additional and different functionality that differentiates one AT
from another, for which no equivalents exist.

3.4 Different choices for the special keys may depend on the operating 

3.5 Conflicts in assignment among keys for different natural languages.

4. Recommendations:

If someone were able to identify the largely common command/control
functionality handled through keyboard combinations among AT software
applications, and indicate which functions are common among them, 
that would be a big help.

4.1 User agents should provide keyboard remapping facilities.

4.2 User agents should provide starter profiles for personalized remappings
from other AT software, and explain differences.

Regards/Harvey Bingham
Received on Wednesday, 7 April 1999 20:11:34 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 20:38:21 UTC