XHTML 2 Draft Recommendation: the @key attribute

I would like to add my support to the arguments John 
Foliot has raised on:

http://www.wats.ca/articles/xhtmlroleaccessmodulestillflawed
/80

In the notes accompanying 

http://hades.mn.aptest.com/cgi-bin/xhtml2-
issues/Role?id=7809

it has been suggested that:

"Author-defined key bindings are a requirement of many 
members of our user community."

I can only conclude that the author's 'user community' 
differs greatly from the user community that I have worked  
with. I have personally *never* encountered a single user 
who was in favour of such author-defined key bindings - let 
alone felt that they were a user requirement. Most feel 
that these bindings are forced upon users (whether they 
want them or not) in a manner that often completely 
disregards the serious conflicts they can cause.

I should like to take this opportunity to point out that, 
for some years, I actually favoured such an approach under 
the mistaken assumption that key bindings (accesskeys)  
offered significant navigational assistance for some users. 

However, after listening to the arguments against 
accesskeys, I finally got around to *asking* the users for 
their opinions.

"I don't use them..."

"They vary from site to site, and I'd have to a) learn 
whether a site has them, b) what they are on that site."

"...they would need to offer *additional*
functions to standard keyboard shortcuts or be an 
additional way to access keyboard functions, not be a 
replacement for them."

"The site should be designed in such a way as to not break 
standard keyboard shortcuts."

All of the above arguments against author defined 
accesskeys can be equally applied to the proposed @key 
attribute. Nice idea, in theory but, in reality, at best, a 
complete waste of time and, at worst, a positive and 
significant hinderance to effective web usage.

The notes on http://hades.mn.aptest.com/cgi-bin/xhtml2-
issues/Role?id=7809 also mention:

"The working group agrees that the end users should be able 
to override key bindings, but authors must be able to 
define them."

Does the working group really believe that the average user 

has the technical ability to over-ride key bindings? Must 
users of assistive software now have to increase their 
learning curve yet further by learning how to dismantle the 
barriers created by authors?

And since when did the words "author" and "must" belong 
together in a discussion about possibly over-riding the 
default behaviour of *users'* software?

Until web authors understand that they must fit in with 
users and not vice versa, large sections of the potential 
user community will remain disenfranchised by a medium that 
has tremendous potential to empower them. The 
implementation of @key will simply continue this rather 
shabby tradition.

I would earnestly request that the working group seriously 
reconsider.

Mel Pedley

-- 
Administrator
Guild of Accessible Web Designers
mel.pedley@gawds.org
http://www.gawds.org

Received on Friday, 18 November 2005 09:59:16 UTC