W3C home > Mailing lists > Public > public-webapps@w3.org > January to March 2013

Re: Keyboard events for accessible RIAs and Games

From: Кошмарчик <garykac@chromium.org>
Date: Wed, 30 Jan 2013 09:10:56 -0800
Message-ID: <CAGnkXoGAVL=QiZWm_uqPQSOhqtpnMRhkPiStZ9w7p=9Sjc4jEg@mail.gmail.com>
To: Florian Bösch <pyalot@gmail.com>
Cc: Webapps WG <public-webapps@w3.org>
This is not the first time these issues have been raised. The following
threads can provide some context to the current state of the discussion:
http://lists.w3.org/Archives/Public/www-dom/2012JulSep/0103.html (obsolete

Please also take a look at the UIEvents (formerly "DOM Level 4 Events")
Editor's Draft at:
I believe that it addresses all of your concerns but I'm interested in any
comments that you might have.


On Wed, Jan 30, 2013 at 6:05 AM, Florian Bösch <pyalot@gmail.com> wrote:

> I have written a blog post at length about this issue here:
> http://codeflow.org/entries/2013/jan/30/keyboard-events-in-javascript-are-broken/
> In short the problem is the following:
> - RIAs (shortcut systems) and Games (shortcut systems and action systems)
> need to be able to identify each key uniquely regardless of modifier keys
> pressed.
> - To make things mappable an unmodified primary symbol or description
> needs to be available to display to the user.
> - To make things speedy to handle there should also be an accompanying
> unmodified keyCode (numeric)
> The current event model (keyCode, charCode and which) do not satisfy that
> requirement:
> - The same keyCode appears on multiple keys
> - The same charCode appears (differently from keyCode) on multiple keys
> - charCode and keyCode change based on modifiers
> - keyCodes cannot be converted to a users locale
> - no non printable key descriptions are offered
> - "which" is a combination of all the problems above.
> It is recognized that this is not a good way to do things as for instance
> the MDN already lists "keyCode", "charCode" and "which" as deprecated.
> The DOM Event Level 3 does not solve all these problems:
> - The char member is still being delivered modified
> - The key member has to match the char member for printable keys
> - There is no numeric unmodified keyCode
> I would suggest:
> - Add a primary key symbol (printable, from the users locale) umodified
> and for non printable keys add the human readable description ('backspace',
> 'enter' etc.)
> - Add an additional unmodified key code member holding the numeric key
> code unmodified
Received on Wednesday, 30 January 2013 17:11:23 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 18:13:58 UTC