Firefox/Goals/2010Q2: Difference between revisions

Jump to navigation Jump to search
simplified
No edit summary
(simplified)
Line 1: Line 1:
{{Draft}}
*{{new|Theme Refresh}}: No blocking bugs for Windows; no beta blocking bugs for Mac or Linux
*New Firefox Theme
**Windows complete (no blocking bugs)
**Beta quality for Mac, Linux


*Feature Work
*{{new|Self hosted home tab and app tabs shipped in a beta}}
**This is the quarter where a bunch of our awesome ideas need to very quickly get very solid.
**Self-hosted Home Tab
**App Tabs
**Design/Implement better approach for privacy/security sensitive notifications


*Jetpack Enablement
*{{new|Geo UI renovated, shipped in a beta}}
**As with our feature work, this is the quarter where Jetpack gets real.
**This isn't "a labs project," this is the Firefox SDK.
**All 0.2-targetted JEPs completed, plus page-worker which is post-0.2 but feels very high value


*Integrated Developer Tools
*{{new|Jetpack: All milestone 0.2 JEPs completed (plus page-worker)}}
**Beta quality designer-friendly inspector
**Beta quality activity console with jetpack integration
**Basic content debugging


*Performance
*{{prev|Developer Tools:}} Inspector and activity console with jetpack integration shipped in a beta. Support for basic content debugging.
**We have some really solid knowledge about Ts and bugs in the queue, but we clearly have problems estimating the amount of ground we can cover in a quarter, too.
**Ts still matters, but we're running into the fact that we don't "believe" clean Ts very much.
***We should refine/rebuild that test to reflect a more plausible startup scenario (e.g. 5 tabs, timer only stops after we: start up, accel-L, type www.locallycachedgoogle.com, load that page)
**Produce a complete startup timeline.
***Identify 5 most expensive modules ("chunks of code"), focus perf work there
***Example: make sessionrestore O(1) within the startup path (!)


===Eng-wide Goals we should expect to be heavily involved with:===
*{{new|Produce a complete startup timeline}}, use it to identify 5 most expensive modules. Set a performance improvement target for those modules by May 1.
 
*Ship Lorentz
**We have work here around making sure the OOPP UI lands safely, but this is really eng-wide, and mostly on bsmedberg right now
 
*Electrolysis
**Able to build, run Firefox with a separate content process. (Things might break, chrome might not work fully, but the browser should stay standing).
 
*J-PAKE
** Working with Shaver, Lucas on how to articulate an eng-wide "land security features" goal
Confirmed users
1,349

edits

Navigation menu