Re: [g-a-devel] Re: Enable accessibility by default in development releases?
- From: "Padraig O'Briain" <Padraig Obriain Sun COM>
- To: Federico Mena Quintero <federico ximian com>
- Cc: gnome-accessibility-devel gnome org, "Derek Rafter Sun COM" <Derek Rafter Sun COM>, John Rice <John Rice Sun COM>, desktop-devel-list gnome org, David Malcolm <dmalcolm redhat com>
- Subject: Re: [g-a-devel] Re: Enable accessibility by default in development releases?
- Date: Tue, 29 Nov 2005 08:16:58 +0000
Our accessibility QA guys have tested these changes and find that they
do not impact the accessibility of gnome-terminal.
My reading of the vte code was that these signals were being emitted too
often. The signal will now be emitted once for every time
vte_terminal_process_incoming rather than at least twice per character.
On Thu, 2005-11-24 at 10:15 -0600, Federico Mena Quintero wrote:
> On Thu, 2005-11-24 at 09:32 +0000, John Rice wrote:
> > Great stuff Padraig - down from +500% to 6% degradation when cat'ing a
> > large file sounds pretty good to me :)
> > Anyone any idea of the effect of not flushing the pending events?
> > Would it cause problems when typing quickly in the terminal? Problems
> > in asian locales?
> The patch removes the "emit those signals" bit.
> I don't know how vte works internally, but it would be good to research
> - Does proper a11y depend on those signals?
> - If it does, then how often do those signals really get emitted? Once
> per character? What happens when you do "ls -l" - does a screen reader
> read out everything to you?
> - Can the a11y implementation for vte throttle those signals, and not
> re-fetch the contents unless the signal has not been emitted for N
> milliseconds? Think of the "ls -l" case above.
> Gnome-accessibility-devel mailing list
> Gnome-accessibility-devel gnome org
] [Thread Prev