Agenda

Should we continue having the 07:00/15:00 meeting time split?
Bastien updated the Universal Access Settings and some orca tips. see https://bugzilla.gnome.org/show_bug.cgi?id=638205
Mattias Clasen gail revamp
AT registration on at-spi2?
Miscellaneous time

Meeting Log

The 07:00 UTC meeting time was created for people mainly in China and Australia.
The people for whom that meeting time is intended, are rarely able to join us at that time.
Having the meeting at the same time every week would be easier to remember.
Li indicated that he should be able to make the 15:00 meeting time.
Mike pointed out that we should be willing to reconsider the decision in the future as team members change. The team concurred.
Joseph offered to make the meeting at 14:30 rather than 15:00, i.e. for Li and others for whom the 07:00 meeting was intended.
Li indicated the slightly earlier time would work well for him.
Other team members agreed that the earlier time would be acceptable for them.

Bastien made some changes to Universal Access Settings and Orca. See 638205 and 642592.
Now the gsetting that says if a11y is enabled is toolkit-accessibility.
This new setting is intended to become activated if you enable the screen reader.
In the new universal access settings there is no longer a "enable accessibility checkbutton."
Bastien also provided a patch (committed) so that if the screen reader is enabled in the universal access settings, Orca is started.
Mike asked where the settings were in GNOME 3. Piñeiro responded in the Universal Access Settings Dialog which you can access via gnome-shell or via the classic desktop.
Joseph asked if there is a reason to turn off accessibility once it is on as there did not appear to be a dialog to turn it off.
Piñeiro responded that as far as he understood it, accessibility will be turned off (i.e. the registry will quit) if it is not required.
Li asked about the atk bridge and gail. Piñeiro responded that this change doesn't involve either; that is currently done by the .desktop files.
Piñeiro added that the only thing directly using this gsetting is gnome-shell.
Piñeiro also stated that there are a lot of doubts about the integration part of the accessibility support.
Ale asked if the gnome-shell accessibility support was documented on the wiki, blogs, etc.
Piñeiro responded that he plans to make a blog post soon.

Piñeiro reported that Matthias was testing gail and discovered some issues. This led to him stating that a general review of the current situation is needed. Matthias wants to take this opportunity to move gail into gtk itself. Mattias has an experimental branch (drop-gail)
Li responded that he is working on this and has committed a patch to move gailarrow to gtkarrow yesterday.
Piñeiro asked about Mattias' statement that it will be required to review part of the focus management code, presumably in gail.c: Is Mattias planning on doing this now or after the move?
Li believes it will be after the move.
Piñeiro asked where the gail.c code will be placed or if gail.c will just be moved.
Li was not sure but thought it would be a gtkmodule, with a file for initialization. But he doesn't know where it will be kept.
Piñeiro asked if Li will be able to continue working on this effort.
Li responded that he believed he would as long as "nothing with higher priority comes out."
Li added that this should give us the opportunity to solve many very old gail bugs, which he had filed even before he was the maintainer.
Mike mentioned that this will hopefully help with treeview performance, adding that Mattias implied that it might.
Li responded that this move creates opportunities to fix things, but much more work will still be needed.

Piñeiro sent a mail to gnome-accessibility-devel
Li stated that although he thinks it is a good idea, it is something that we need to check.
Joanie stated that it sounded like a good idea, but she is leaving it up to people like Li and Mike to comment on its feasibility.
Li responded that we can try it.
Piñeiro pointed out that most people do not require accessibility support.
Mike asked if Piñeiro was proposing we make this change for GNOME 3.0.
Piñeiro responded that this was his proposal; rather than make a bug to be added on to the "towards atk2.0" work.
Mike stated that right now AT-SPI2 shouldn't send events over the wire if no one is listening, but it will still listen for gail events.
Piñeiro replied that he wasn't talking about at-spi2, but the bridge which sends the gail (atk implementor) events to at-spi2.
Mike stated that we'd then load gail and atk-bridge all the time, and only really enable atk-bridge when accessibility is enabled
Piñeiro asked if Mike was proposing to just avoid the overhead if accessibility is enabled instead of a AT registration method?
Mike responded that this was his interpretation of Piñeiro's proposal and that it was not clear to him how else we could have it register.
The discussion was concluded by Piñeiro on the grounds that the meeting was not the place to discuss technical details.

Joseph reported that Magnifier settings have been moved outside of gnome-shell and that the Universal Access Settings will enable magnifier configuration.
Trev reported that Fer asked about an API break for Atk 638377 so that Mozilla bug #619002 can be fixed.
- Mike asked if the code had already been committed and if it was okay to add it now?
- Trev responded that it has not been committed as that would require breaking freeze.
- Piñeiro replied that it was his understanding that Fer made a change to avoid the ABI break.

The team should test the new Universal Access Settings dialog in a GNOME 3 "compliant" environment.
Piñeiro will create a bug for his AT registration on AT-SPI2 proposal.