headers
remus draica | 4 Feb 2004 11:33
Picon

Re: Selection and Combo Boxes


Hi Padraig,

If user press the up/down arrow key, then a selection-changed event from the 
combo is fired. If combo doesn't implement AccessibleSelection interface, 
what kind of events will be fired in this case?

Regards,
Remus

On Friday 30 January 2004 12:29, padraig o'briain wrote:
> remus draica wrote:
>
> Thanks.
>
> If I read the code correctly, you do not make an assumption that a
> combo-box implements AccessibleSelection. What would happen if a combo-box
> did not implement AccessibleSelection.
>
> Padraig
>
> > Hi Padraig,
> >
> > Gnopernicus uses AccessibleSelection for combos.
> > (srlow/libsrlow/SRLow.c line 1149).
> >
> > Regards,
> > Remus
> >
> > On Thursday 22 January 2004 16:52, padraig o'briain wrote:
(Continue reading)

Permalink | Reply | 3 comments |
headers
Bill Haneman | 4 Feb 2004 12:17
Picon

Re: Re: Selection and Combo Boxes

remus draica wrote:

>Hi Padraig,
>
>If user press the up/down arrow key, then a selection-changed event from the 
>combo is fired. If combo doesn't implement AccessibleSelection interface, 
>what kind of events will be fired in this case?
>
>Regards,
>Remus
>
>
>On Friday 30 January 2004 12:29, padraig o'briain wrote:
>  
>
>>remus draica wrote:
>>
>>Thanks.
>>
>>If I read the code correctly, you do not make an assumption that a
>>combo-box implements AccessibleSelection. What would happen if a combo-box
>>did not implement AccessibleSelection.
>>    
>>
I believe Padraig may be proposing that the selection-changed events 
would come from the "list child" of the combobox, which would implement 
AccessibleSelection.
The intention of AccessibleSelection was that selection takes place on 
its direct children, whereas the list items in a ComboBox are more 
deeply nested.
(Continue reading)

Permalink | Reply | 2 comments |
headers
Peter Korn | 4 Feb 2004 20:19
Picon

Re: Re: Selection and Combo Boxes

Hi Bill,

> I believe Padraig may be proposing that the selection-changed events
> would come from the "list child" of the combobox, which would implement
> AccessibleSelection.
> The intention of AccessibleSelection was that selection takes place on
> its direct children, whereas the list items in a ComboBox are more
> deeply nested.

Yes, that was the intention, though I think "assumption" might be more
accurate.  Given the way the API works, I'm not certain we can easily change
it, but it may be worth rethinking/revisiting.  Should AccessibleSelection
really be about selectable descendents (vs. only immediate children)?  What are
the implications of changing this meaning?

Peter Korn
Sun Accessibility team
Permalink | Reply | 1 comment |
headers
Bill Haneman | 5 Feb 2004 11:27
Picon

Re: Re: Selection and Combo Boxes

Peter Korn wrote:

>Hi Bill,
>
>  
>
>>I believe Padraig may be proposing that the selection-changed events
>>would come from the "list child" of the combobox, which would implement
>>AccessibleSelection.
>>The intention of AccessibleSelection was that selection takes place on
>>its direct children, whereas the list items in a ComboBox are more
>>deeply nested.
>>    
>>
>
>Yes, that was the intention, though I think "assumption" might be more
>accurate.  Given the way the API works, I'm not certain we can easily change
>it, but it may be worth rethinking/revisiting.  Should AccessibleSelection
>really be about selectable descendents (vs. only immediate children)?  What are
>the implications of changing this meaning?
>
I've argued against changing or extending this meaning to Padraig, as it 
makes the client
heuristics much more complex.  It also would cloud the meaning of the 
'index' parameters needed in order to modify selections.

My view is that our current API requires that 'selections' be direct 
children of the selectable object, and that if we want to rethink this 
will will need a different API.
(Continue reading)

Permalink | Reply | 0 comments |
headers
Calum Benson | 5 Feb 2004 16:57
Picon

Is 406 a new Peugeot?

No, it's the number of accessibility bugs we have open today:

8 Stoppers (no change since 27th Jan)	in=6, out=6
50 Candidate stoppers (-5) 		in=12, out=17
218 Significant, not stoppers (+9)	in=52, out=43
130 Lesser impact (-2) 			in=2, out=4

37 of which are as yet untriaged by the accessibility team.

List of open bugs only:
http://www.gnome.org/~calum/access-bugs-openonly.html

Full report:
http://www.gnome.org/~calum/access-bugs.html

(NB: Only bugs affecting modules in Sun's distro are currently included
in the bug counts.)

Cheeri,
Calum.

--

-- 
CALUM BENSON, Usability Engineer       Sun Microsystems Ireland
mailto:calum.benson <at> sun.com            GNOME Desktop Group
http://ie.sun.com                      +353 1 819 9771

Any opinions are personal and not necessarily those of Sun Microsystems
Permalink | Reply | 4 comments |
headers
Luis Villa | 5 Feb 2004 17:54

Re: [g-a-devel]Is 406 a new Peugeot?

<blink>

I think I'd only ever looked at the top couple of lines. 406? When did
this number start expanding so radically? There were less than 300 two
months ago. Is there some particular phenomenon causing this? Are we
getting more sloppy in our coding? Testing getting more thorough?

Luis

On Thu, 2004-02-05 at 10:57, Calum Benson wrote:
> No, it's the number of accessibility bugs we have open today:
> 
> 8 Stoppers (no change since 27th Jan)	in=6, out=6
> 50 Candidate stoppers (-5) 		in=12, out=17
> 218 Significant, not stoppers (+9)	in=52, out=43
> 130 Lesser impact (-2) 			in=2, out=4
> 
> 37 of which are as yet untriaged by the accessibility team.
> 
> List of open bugs only:
> http://www.gnome.org/~calum/access-bugs-openonly.html
> 
> Full report:
> http://www.gnome.org/~calum/access-bugs.html
> 
> (NB: Only bugs affecting modules in Sun's distro are currently included
> in the bug counts.)
> 
> Cheeri,
> Calum.
(Continue reading)

Permalink | Reply | 2 comments |
headers
alonzo cuellar | 5 Feb 2004 18:06
Picon
Favicon

speech sounds like a chickmunk

Hello there all.
I have installed Gnopernicus and have the festival
speech software synthisizer running.  However, the
speech sounds like a chickmunk, I have layer eight and
decreased the pitch of the speech.
it still sounds the same.  I also run the say program
for dectalk and it also sounds highly pitched.
Does anybody know what is the matter?

__________________________________
Do you Yahoo!?
Yahoo! Finance: Get your refund fast by filing online.
http://taxes.yahoo.com/filing.html
Permalink | Reply | 0 comments |
headers
Calum Benson | 5 Feb 2004 18:37
Picon

Re: [g-a-devel]Is 406 a new Peugeot?

On Thu, 2004-02-05 at 16:54, Luis Villa wrote:

> I think I'd only ever looked at the top couple of lines. 406? When did
> this number start expanding so radically? There were less than 300 two
> months ago. Is there some particular phenomenon causing this? Are we
> getting more sloppy in our coding? Testing getting more thorough?

Testing is certainly contributing to it; our QA guys are running several
a11y scenarios every week now with the 2.5 stack and filing a fair
number of bugs from that.  I can't comment on the sloppiness or
otherwise of anyone's coding though :)

Cheeri,
Calum.

--

-- 
CALUM BENSON, Usability Engineer       Sun Microsystems Ireland
mailto:calum.benson <at> sun.com            GNOME Desktop Group
http://ie.sun.com                      +353 1 819 9771

Any opinions are personal and not necessarily those of Sun Microsystems
Permalink | Reply | 0 comments |
headers
remus draica | 6 Feb 2004 13:29
Picon

A major change in gnopernicus


Hi all,

Because gnopernicus uses now XML files (instead gconf keys) to store 
presentation information (what should be present for an obkect in case of an 
event) and a translation was made from previous format, please report any 
difference in what is presented for an object from previous version.

Best regards,
Remus
Permalink | Reply | 2 comments |
headers
Bill Haneman | 6 Feb 2004 17:45
Picon

Re: A major change in gnopernicus

remus draica wrote:
> 
> Hi all,
> 
> Because gnopernicus uses now XML files (instead gconf keys) to store
> presentation information (what should be present for an obkect in case of an
> event) and a translation was made from previous format, please report any
> difference in what is presented for an object from previous version.

Just to clarify:  The intention is that this change does _not_ change what is presented to the user.  But since
the implementation backend did change significantly (in order to allow proper translation), there is
the possibility that things have indeed changed.  I believe Remus is letting everyone know the intention
so that any such accidental changes can be fixed in a timely fashion.

best regards,

- Bill

> Best regards,
> Remus
> _______________________________________________
> Gnome-accessibility-devel mailing list
> Gnome-accessibility-devel <at> gnome.org
> http://mail.gnome.org/mailman/listinfo/gnome-accessibility-devel

--

-- 
------
Bill Haneman 
Gnome Accessibility Project
Sun Microsystems Ireland
(Continue reading)

Permalink | Reply | 0 comments |

Gmane