Re: Issue #69: Character key commands

> can someone help me here? 

Hi Gregg,
I'll give a specific example. It's not using google sites, it's using 
gmail, where the  "e", is not a shortcut to Edit, but to Archive a 
message.

I am using google email. I think my focus is in the edit field of the 
message I'm reading, so I begin to type a response to the user. The first 
word I begin to type is "Every". Since my focus is not actually on the 
edit field, google immediately archives the entire message and displays a 
little bubble that reads "The conversation has been archived.  Learn more
  Undo". But I'm typing, so this message disappears immediately when I 
type the "v" and then when I type the second "e" gets updated to read "No 
conversations selected." which persists through the rest of the letters 
("ry") so that I actually see that message if I'm paying attention. (Such 
messages do not persist indefinitely; not sure what their algorithm is).

The focus cursor gets out of the edit field in gmail more frequently than 
you would think. I'm not sure what actions I do that cause this, but  I 
accidentally archive messages frequently.

I completely agree that single key shortcuts can be highly beneficial. But 
I equally agree that giving the user the option to disable them is a 
valuable feature, because while it is very simple to accidentally trigger 
one (as I hope I have shown), the same is not true of a non-printable 
character or a combo that involves a modifier key. I could accidentally 
hit the former, but that is a much less common problem than intentionally 
triggering a printable key in the wrong context. The modifier key combo is 
even more 'safe' since the odds of accidentally hitting both a modifier 
and a regular key at the same time are pretty low for most circumstances.

Michael Gower
IBM Accessibility
Research

1803 Douglas Street, Victoria, BC  V8T 5C3
gowerm@ca.ibm.com
voice: (250) 220-1146 * cel: (250) 661-0098 *  fax: (250) 220-8034



From:   Gregg C Vanderheiden <greggvan@umd.edu>
To:     Greg Lowney <gcl-0039@access-research.org>
Cc:     "w3c-waI-gl@w3. org" <w3c-wai-gl@w3.org>
Date:   2017-05-15 10:46 PM
Subject:        Re: Issue #69: Character key commands



agree

however - I have a question?

Google does this on Google Site pages.

when you are NOT in edit mode — you can type the “e” key to go into edit 
mode.
when not in edit mode there is nothing else that the ‘e’ key can be used 
for.    It will simply be eaten by the page and have no effect of any 
kind.

What is the accessibility problem caused by this? 
 I can see how it is very convenient to a person with a disability
 (if they saved and are scrolled down the page - hitting ‘e’ puts them 
into edit mode where they are rather than having to scroll all the way to 
the top, hit the edit button, the scroll all the way back down to find the 
location where they were. 
it also prevents them from having to do a choric keyboard input to get it 
back into edit mode. 

But I can’t think of the disadvantage. 

can someone help me here? 
it is a ‘hidden’ command but so are all the two key combinations
it prevents the user from using the ‘e’ key for anything else — but (since 
this only happens if they have focus on the page — and there is nothing 
that the ‘e’ key can do on the page when not in edit mode — I’m missing 
the problem.


thx

g 

Gregg C Vanderheiden
greggvan@umd.edu




On May 16, 2017, at 2:33 AM, Greg Lowney <gcl-0039@access-research.org> 
wrote:

The latest draft wording as of 2017-05-15 reads "Character Key Shortcuts: 
If a keyboard shortcut consisting entirely of character keys is 
implemented by the web page, then a mechanism is available to turn it off 
or to remap it to a shortcut that uses at least one non-printing key. 
(Level A)".

On further reflection, I see no reason to limit this SC to keyboard 
shortcuts implemented by content rather than have it apply more broadly to 
all keyboard commands implemented by content. That is, if a script traps 
the F key and uses it to toggle between normal and full-screen modes, it 
is equally bad for the user whether it does so by activating a button on 
the page that toggles the mode (making it a shortcut) or merely toggling 
the mode independently of any controls (in which case it's not a 
shortcut). Generalizing it also lets us avoid having to define keyboard 
shortcuts.

However, if the keyboard command is the only way to carry out a action, 
then it is not acceptable for the content to merely disable the keyboard 
command. We can address this by adding the caveat "without loss of content 
of functionality" we're using elsewhere. In fact, this should probably be 
added in any case.

I also have three purely editorial comments: (1) We're supposed to use the 
term "content" rather than "web page". (2) We should standardize on either 
"character key" or "printing key" rather than mixing the two terms in the 
same to mean the same thing. While "printing key" is the one I'm more used 
to seeing in the past, I think "character key" is more intuitive to 
readers unfamiliar with the concept. (3) Although it may be overly 
pedantic, "entirely of character keys" actually introduces ambiguity as to 
whether the plural implies it must be more than one character key. To 
avoid that we could say "consisting entirely of one or more character 
keys" or "that does not include any non-character keys".

Incorporating the editorial changes alone would change it to read: 
"Character Key Shortcuts: If a keyboard shortcut consisting entirely of 
one or more character keys is implemented by the content, then a mechanism 
is available to turn it off or to remap it to use at least one non-
character key. (Level A)"

Incorporating the change to keyboard commands alone would change it to 
read: "Character Key Commands: If a keyboard command consisting entirely 
of character keys is implemented by the web page, then a mechanism is 
available to turn it off or to remap it to a command that uses at least 
one non-printing key, without loss of content or functionality. (Level 
A)".

Incorporating both sets of changes would change it to read: "Character Key 
Commands: If a keyboard command consisting entirely of one or more 
character keys is implemented by the content, then a mechanism is 
available to turn it off or to remap it to use at least one non-character 
key, without loss of content or functionality. (Level A)"

    Thanks,
    Greg