Meeting Summary

Meeting Start Times, Quorum, and Waiting for People

When we don't have too many people at the meeting, we wonder about cancelling the meeting, or just doing a short meeting.
But in most of the cases, we waited like 30 minutes for people.
This happened recently, mostly due to Christmas holidays.
Last week we did that, we started 30 minutes late, but in the end the meeting was longer that expected and we finished more than 30 minutes late. And that was not the first time this has happened.
Taking into account that starting 30 minutes late, and ending 30 minutes late is somewhat silly, from now on, if there are people when the meeting starts, and someone willing to lead the meeting, we are going to start the meeting on time (or at least, just waiting 5 minutes instead of 30).
If meeting finish in 10 minutes, that would be ok, people would be free to go out without being 30 minutes hanging around on IRC.
If we do not have enough people for a quorum, we will discuss but not make binding decisions at the meeting in question.

At the end of last week, Joanie did a little bit of investigation on making JAW build.
It was a slow process of make a change, try again, google, read docs, rinse and repeat.
Joanie thinks that given the lack of response from everyone else, perhaps she and Mike and Piñeiro and *maybe* Javier Jardon can get this problem solved.
Joanie mentions Javier because he has autotools skills (she thinks).
Once the blessed thing builds, then we can start debugging and get things re-working; or make a more educated judgment that it's not worth the effort.
Joanie will, as her spare time permits, try to get it building. She hopes Mike can help out too.

Joanie filed bugs for the Accerciser Python 3 issues she found and attached patches.
Javi just returned from vacation and has not had time to review them yet.

Last week we didn't have too many people at the meeting, but we had a long update.
About GNOME 3.8 in relation with webkit2, please read the log.
Benjamin Otte announced some changes on Gtk related to not exposing widgets which are not visible. He and Mike Gorse are working on that.
Benjamin sent a mail to gtk-devel announcing just in case.
Piñeiro responded adding some info and asking some questions.
Eventually, it would be good to do this change for clutter.
Mike has been struggling to find the memory leaks in pygobject but doesn't understand the code very well.
Seems we are not getting keyboard events from gnome-control-center, though Joanie isn't sure if it's a clutter-gtk thing or something else. See bug 691468.

Juanjo started an article about a11y in Spanish for GNOME Hispano (Spanish-speaking GNOME association).
The article is mainly a translation of the Linux.com artcile with some minor updates/changes.

Joanie:
- Do a full evaluation of the accessibility breakage in Evolution 3.6 and file bugs accordingly.
Juanjo:
- Take a look at the current a11y material in order to make an action plan.
Piñeiro:
- Create some usability-configurability related bugs, using Bryen's mail as a starting point.
- Investigate that "Enable by Keyboard" thing.
- Update the AtkObject:children-changed documentation, as this could be also used on ATK implementors.