W3C home > Mailing lists > Public > wai-xtech@w3.org > August 2007

Re: argument for only ONE set of radio button navigation keys

From: David Poehlman <david.poehlman@handsontechnologeyes.com>
Date: Sat, 25 Aug 2007 08:58:28 -0400
Message-ID: <005b01c7e717$a2b31d60$0601a8c0@HANDS>
To: <wai-xtech@w3.org>, "Pete Brunet" <brunet@us.ibm.com>

pete, user learns from feedback, if they find that left and right don't 
work, they quickly find that up and down do and they can see it happen 
visually.  Unfortunately, the very real issue that Gregory faces is not well 
documented but is no less important because of that.  The fewer keystrokes 
one needs to use to do something the better.  perhaps we need a stipulation 
that this can be altered by user preference.

----- Original Message ----- 
From: "Pete Brunet" <brunet@us.ibm.com>
To: <wai-xtech@w3.org>
Sent: Friday, August 24, 2007 6:08 PM
Subject: Re: argument for only ONE set of radio button navigation keys


>up and down, left and right are all in the eye of the beholder,
>and the orientation of characters in a written language -- that
>is why, as a user, i would rather have ONE set of keys defined
>for radio button behavior -- regardless of visual orientation --
>and one set of keys reserved for expansion and collapsing tree
>views...  it's a question of seperating presentation from
>content, and reducing the burden on the user...
Hi Gregory, I don't understand the burden.  It was years before I
realized, as a sighted user, that I could use either pair.  It was never a
burden to not know of the missing pair.

>besides, those
>with neuropathy and any other tactile disabilities (such as
>myself, he quickly added in an attempt at "full disclosure")
>i would rather have a 4 key arrangement limited to 2, as i
>don't want to move selection just because i can't feel which
>key is which, and if i make an error (using leftarrow instead
>of up arrow, i don't want the widget making decisions for me
>which i never intended to communicate to it
That might be a good argument for your position.  I've not heard of this
issue being raised before.  If this is a issue, what techniques are used
to resolve the problem?  Would those techniques serve just as well in the
case we are discussing?  Is there data from the a11y UX community that can
be used to bolster your position?

I have often seen surveys using groups of radio buttons arranged
horizontally.  In that case not having left/right would be confusing for a
sighted user.

>when navigating a tree i'd rather
>have only 2 responsive keys than 4...
Isn't tree behavior a separate issue?  In any event, I don't see how you
can have less than 4 for trees.  When using up/down I wouldn't want nodes
to open.  This would be very cumbersome for deep trees.  When using
left/right I think users expect both open/close and traversal, i.e. using
right on a closed node opens the node, right again moves to the first
child.

Pete Brunet

IBM Accessibility Architecture and Development
11501 Burnet Road, MS 9022E004, Austin, TX 78758
Voice: (512) 838-4594, TL 678-4594, Fax: (512) 838-9666
Ionosphere: WS4G




"Gregory J. Rosmaita" <oedipus@hicom.net>
Sent by: wai-xtech-request@w3.org
08/24/2007 03:29 PM

To
Becky Gibson/Westford/IBM@Iris, wai-xtech@w3.org
cc

Subject
Re: argument for only ONE set of radio button navigation keys







becky wrote, quote:
I think you are suggesting that only one set of arrows be implemented -
either up/down and left/right.
unquote

yes, that is precisely what i'm suggesting...

becky also wrote:
My feeling that supporting both was accommodating to all users.  I
believe that a visual l person encountering a horizontal grouping
would assume left and right key navigation and up and down key
navigation for a vertical orientation.  By supporting both sets the
person who can not perceive the orientation can use either set of
keys.
unquote

ah, but this is a problem that the XHTML2 working group has had
to address with what used to be called a "horizontal rule" -- 
for languages that are vertically aligned, horizontal rules don't
help much, and it is to keep authors from using kludges that the
redefined seperator (similar to the LS i proposed last year and
also submitted to the HTML WG for consideration) alleviates...

up and down, left and right are all in the eye of the beholder,
and the orientation of characters in a written language -- that
is why, as a user, i would rather have ONE set of keys defined
for radio button behavior -- regardless of visual orientation --
and one set of keys reserved for expansion and collapsing tree
views...  it's a question of seperating presentation from
content, and reducing the burden on the user...  besides, those
with neuropathy and any other tactile disabilities (such as
myself, he quickly added in an attempt at "full disclosure")
i would rather have a 4 key arrangement limited to 2, as i
don't want to move selection just because i can't feel which
key is which, and if i make an error (using leftarrow instead
of up arrow, i don't want the widget making decisions for me
which i never intended to communicate to it -- in radio button
grouping only 1 set of arrow keys should work -- the others
should do nothing; likewise, when navigating a tree i'd rather
have only 2 responsive keys than 4...

i hope this addresses your question,
gregory
-------------------------------------------------------
BRAIN, n.  An apparatus with which we think we think.
              -- Ambrose Bierce, The Devil's Dictionary
-------------------------------------------------------
Gregory J. Rosmaita, oedipus@hicom.net
Oedipus' Online Complexes: http://my.opera.com/oedipus/
-------------------------------------------------------
Received on Saturday, 25 August 2007 12:58:35 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 27 April 2012 13:15:43 GMT