- From: Chris Blouch <cblouch@aol.com>
- Date: Wed, 19 Dec 2007 14:40:23 -0500
- To: John Foliot <foliot@wats.ca>
- CC: 'Aaron M Leventhal' <aleventh@us.ibm.com>, 'W3C WAI-PFWG' <w3c-wai-pf@w3.org>, w3c-wai-pf-request@w3.org, 'W3C WAI-XTECH' <wai-xtech@w3.org>, wai-xtech-request@w3.org, 'HTMLWG' <public-html@w3.org>
- Message-ID: <476973A7.9030702@aol.com>
As an implementation detail in our approach at AOL, our keyboard shortcuts are instrumented not only on a per-page basis but can also be assigned to objects in the DOM. So the onus is on the UI/experience designer to maintain some level of consistency between different pages on a web application. The DOM level assignment is quite handy so that some commands are contextual to which object currently has focus. For example, control+alt+r reads the mail message currently focused and when focus is not on a mail message the key combination is inert. This allows some consistency of mapping behaviors to keystrokes, much like copy and paste. We don't care what you are copying. Just copy whatever the user currently has given their attention to (selected/focused). CB John Foliot wrote: > Aaron (and others), > > Back when XHTML2 WG was discussing @role, I tried very hard [ > http://www.wats.ca/show.php?contentid=47 ] to have them exclude the > @key attribute in favor of placing the "discoverability" of these > types of enhancements on the shoulders of the user-agent and user. > Given the multitude of possible web-apps that conceivably will emerge > all wanting to map to keyboard accelerators, there will always be a > "battle" - if web app A, web app G and web app W all want to map to > [accelerator] + 6 how will this be resolved? To my mind, clearly, two > of the 3 web app developers will lose out. (This does not even begin > to also consider the i18n issue...) > > On the other hand, if a user-agent, or combination of user-agent and > assistive technology are "alerted" to the existence of supplemental > keyboard accelerators 'embedded' into the app/page/plug-in, and then > allows the *user* to map a key to that 'point', then it would seem (to > me anyway) that everyone is happy: web app developer, *and* user. > Already we sort of have this type of functionality within Opera - user > customization of keystrokes. > > In the interest of true accessibility and universality, it seems this > should be a no-brainer... (at least to me) > > JF > > > > -----Original Message----- > *From:* wai-xtech-request@w3.org [mailto:wai-xtech-request@w3.org] > *On Behalf Of *Aaron M Leventhal > *Sent:* December 19, 2007 9:48 AM > *To:* Wlodkowski, Thomas > *Cc:* Jon Gunderson; W3C WAI-PFWG; w3c-wai-pf-request@w3.org; W3C > WAI-XTECH; wai-xtech-request@w3.org > *Subject:* RE: Reserved keystrokes for borwsers and operating > system functions > > > BTW we have the same issue with plugins. > > Right now when they have focus they steal *all* keystrokes and the > browser gets none. > > There was a developer suggesting we change things around and only > allow content/plugins to get keystrokes we don't absolutely need. > It brings up a lot of concerns, but I'm not sure I'd rule out that > approach entirely. It needs to be considered. > > - Aaron > > > > > *"Wlodkowski, Thomas" <Thomas.Wlodkowski@corp.aol.com>* > Sent by: w3c-wai-pf-request@w3.org > > 12/19/2007 12:24 PM > > > To > Aaron M Leventhal/Cambridge/IBM@IBMUS, "Jon Gunderson" > <jongund@uiuc.edu> > cc > "W3C WAI-PFWG" <w3c-wai-pf@w3.org>, "W3C WAI-XTECH" > <wai-xtech@w3.org>, <wai-xtech-request@w3.org> > Subject > RE: Reserved keystrokes for borwsers and operating system functions > > > > > > > > > > The battle between widgets and browser access to the keyboard is > why the AXS library developed by Chris Blouch on my team now uses > Alt+Control+the letter key. > > To be honest, I'm not sure how the keyboard battle is resolved if > there isn't some path toward better cooperation between the widget > and browser. My company develops web apps that in many cases > replace software applications. I suspect consumers want the same > or equivalent behaviors when they're accessing email via a webmail > app. Hopefully this issue will spark greater participation in the > DHTML Style Guide working group that meets weekly on Tuesdays at > noon. You can see the widget-specific keyboard behaviors discussed > thus far at: > _http://dev.aol.com/dhtml_style_guide_ > > Tom > > > ------------------------------------------------------------------------ > > *From:* wai-xtech-request@w3.org [mailto:wai-xtech-request@w3.org] > *On Behalf Of *Aaron M Leventhal* > Sent:* Tuesday, December 18, 2007 3:31 PM* > To:* Jon Gunderson* > Cc:* W3C WAI-PFWG; W3C WAI-XTECH; wai-xtech-request@w3.org* > Subject:* Re: Reserved keystrokes for borwsers and operating > system functions > > > I have an old list but it should be updated. Could be a great > resources if it was. > http://www.mozilla.org/access/keyboard/ > > Anyway, in the case of Alt+left it didn't appear the grid really > meant to use it. Better to check always the modifier key and let > it go if it's not the key you're looking for. > > - Aaron > > > *Jon Gunderson <jongund@uiuc.edu>* > > 12/18/2007 03:27 PM > > > To > Aaron M Leventhal/Cambridge/IBM@IBMUS > cc > W3C WAI-PFWG <w3c-wai-pf@w3.org>, W3C WAI-XTECH > <wai-xtech@w3.org>, wai-xtech-request@w3.org > Subject > Reserved keystrokes for borwsers and operating system functions > > > > > > > > > > > > Aaron, > Is there a list of key combinations that ARIA (Web 2.0) > applications should never use? > > There are already some conflicts in the best practices, in that > case the best practices says the Widget should win. > > Are these combinations going to be OS and browser specific? > > Jon > > > ---- Original message ---- > >Date: Tue, 18 Dec 2007 14:53:54 -0500 > >From: Aaron M Leventhal <aleventh@us.ibm.com> > >Subject: Re: Updated Illinois ARIA Examples > >To: Jon Gunderson <jongund@uiuc.edu> > >Cc: W3C WAI-PFWG <w3c-wai-pf@w3.org>, W3C WAI-XTECH > <wai-xtech@w3.org>, wai-xtech-request@w3.org > > > > I see a bug in the grid example. > > > > If I hit Alt+left or Alt+right to move in my browser > > history, it moves in the grid instead. The script is > > stealing those keystrokes instead of letting it go > > to the browser. > > You have to return true from your event handler if > > the key was not used. That lets it bubble to the > > next possible handler. > > > > - Aaron > > > > Jon Gunderson To W3C WAI-PFWG > > <jongund@uiuc.edu> <w3c-wai-pf@w3.org>, > > Sent by: W3C WAI-XTECH > > wai-xtech-request@w3.org <wai-xtech@w3.org> > > cc > > 12/17/2007 05:46 PM Subject Updated Illinois > > ARIA Examples > > > > > > I have update the following examples to use the > > non-namespaced version of the ARIA markup: > > > > Menubar > > http://test.cita.uiuc.edu/aria/menubar/ > > > > Grid > > http://test.cita.uiuc.edu/aria/grid/ > > > > Slider > > http://test.cita.uiuc.edu/aria/slider/ > > > > Slider > > http://test.cita.uiuc.edu/aria/slider/ > > > > Tab Panel > > http://test.cita.uiuc.edu/aria/tabpanel/ > > Jon Gunderson, Ph.D. > > Coordinator of Assistive Communication and > > Information Technology (DRES) > > > > WWW: http://www.cita.uiuc.edu/ > > WWW: https://netfiles.uiuc.edu/jongund/www/ > Jon Gunderson, Ph.D. > Coordinator of Assistive Communication and Information Technology > (DRES) > > WWW: http://www.cita.uiuc.edu/ > WWW: https://netfiles.uiuc.edu/jongund/www/ > > >
Received on Wednesday, 19 December 2007 19:43:07 UTC