Canmove, confirm
432
edits
Changes
→next meeting: Add 10/17 notes
= next meeting =
([https://fx-team.etherpad.mozilla.org/australis-meeting-notes next meeting has not yet occurred])
=10/17/2013=
* Milestone 9
** http://i.imgur.com/0p5Tk.gif
* Perf regression and other landing blockers
** mconley is on PTO today, but these notes are fresh!
*** and so clean, ain't nobody so dope as mconley
*** see also, http://www.conleypackaging.com/
** bug 880611 (ts_paint regression)
*** vladan has told me that while there appears to be still some 5ms or so of regression here not accountable to TabsInTitlebar or CustomizableUI, that we shouldn't block on it (in exchange, he wants me to be his contact for some cold-start perf work he's starting up soon). So this is no longer a landing blocker.
** bug 902024 (TART regression)
*** Bug 921038 - Move selected tab linear-gradient into stroke image and switch to SVG curves
**** [mconley] We're going to try putting the linear-gradient, stroke, and background all into a single SVG image, and get rid of our pseudoelements. This has the potential to fix a number of non-performance bugs.
**** [mconley] With SVG caching, this is going to knock out our Windows XP regression, and should put us in the black there.
**** [mconley] I have a high degree of confidence that this will knock out the (small) Linux regressions we have as well. I'll have a higher degree of confidence once some try builds come back.
**** [mconley] MattN is leading this work.
***** I looked into different approaches to use SVG again and avoid the clip-path and they all have limitations and are risky.
****** Bug 914617 may be less risky but we need to figure out a timeline for it
*** Bug 924415 - Find out why setting chromemargin to 0,-1,-1,-1 is so expensive for TART on UX branch on OS X
**** [mconley] This accounts for pretty much all of the regression on OS X 10.6.
**** [mconley] https://bugzilla.mozilla.org/show_bug.cgi?id=924415#c15 breaks down where we're at.
*** Bug 925413 - Find out why drawing the nav-bar background is 10x more expensive on the UX branch
**** This was solved with bug 925514, but didn't seem to result in any significant improvement.
** ts/tpaint data: https://docs.google.com/spreadsheet/ccc?key=0Asj8iLTl0K0UdDcxemJEWVVXWTlKek1wR09iS2FsOXc
** tart data: https://docs.google.com/a/mozilla.com/spreadsheet/ccc?key=0Asj8iLTl0K0UdGJzVnVYYkJTSDREbndRWWFTLWNyS1E
* Bug tracking
** P1: -1
** P2: -3
** P3: -7
** P4: +0
** P5: -3
*** woo!
***
* Roundtable
** [jaws] Removing the padding around the window when in customization
*** We have a good amount of bugs that would go away if we removed this padding
**** Bug 886444 - [Australis] Can't drag window while in customize mode
**** Bug 873060 - [meta] Make entering and existing customization mode feel smooth
**** Bug 871158 - Missing borders when you're in customization mode and the window is maximized
**** bug 885069 Australis: WinXP: too much padding between tabs and menubar when customizing with sizemode=normal (WFM now)
**** bug 883145 - Window gradient in normal state doesn't line up with window gradient in customize mode
**** bug 885066 - lightweight themes should not get applied to windows already in customization mode
** [bwinton] (No voice update needed)
*** Please add stuff to https://firefox-ux.etherpad.mozilla.org/measuringaustralis :)
* Action Items (Last Week)
* Action Items (New)
** followup with roc/seth on bug 914617 since it sounds like the svg-as-image caching won't help TART, but this one should (and help avoid needing to redo a bunch of tab work)
** shorlander & UX: thinking on alternative transitions for entering/exiting customization mode
=10/10/2013=
