2015 Q4
List of objectives and key results for 2015 Q4.
Overview
- Objective / Key Result Owner(s) Score
- RAIL (Measuring User Experience)
- Ship CSS containment to beta leviw 0.60
- Smooth scrolling for wheel and keyboard scrolls in Win/Linux enabled on trunk skobes, ymalik 1.00
- Intersection observer on trunk szager 0.95
- Integrate font related memory consumption to memory-infra drott 1.00
- Interop (Interoperability, Compatibility, Conformance)
- Have fullscreen implementation not modify layout tree bokan 0.00
- Have flexbox match latest version of spec cbiesinger 0.80
- Work with other vendors to nail down Custom Layout API ikilpatrick 0.90
- Support Blink/Android hyphen design work szager 1.00
- Replace flexbox tests with W3C ones cbiesinger 1.00
- Unprefix CSS Writing modes kojii 0.70
- CSS font-feature-settings kojii 0.00
- Code health
- Have root layer scrolling pass all unit tests skobes, szager 0.50
- Have prototype of ruby on top of custom layout [kojii, houdini-dev] blink-houdini, kojii 1.00
- Remove simple text eae, drott 0.90
- Ship improved font fallback to stable drott 0.75
- Finish wrapping block layout in an API leviw, pilgrim 0.70
- Add class level documentation to all key layout classes jchaffraix 1.00
- Feasibility study for replacing flipped flocks writing mode implementation wkorman 0.00
- Bug Health
- Continue to triage all bugs within one week eae 1.00
- Fix all (new) P0/P1 within one release eae 0.80
- Reduce untriaged bug count by 50% eae, jchaffraix 1.00
- Reduce total bug backlog by 15% eae 1.00
Details
Ship CSS containment to beta
Owner(s):leviw Score:0.60
Editor’s Draft https://drafts.csswg.org/css-containment/ [mid quarter] On track, intent to implement has been sent. Patch up for review. [end of quarter] CSS and Paint containment implemented and available as experimental web platform features. We're further along implementation wise than expected however behind on shipping as CSS contain has yet to ship to beta.
Smooth scrolling for wheel and keyboard scrolls in Win/Linux enabled on trunk
Owner(s):skobes, ymalik Score:1.00
https://crbug.com/575 [mid quarter] On track, looking good. [end of quarter] Enabled on trunk in time for the M49 branch.
Intersection observer on trunk
Owner(s):szager Score:0.95
[mid quarter] On track, harder and more time consuming than estimated. [end of quarter] More work and much harder than originally estimated. Implementation has been ready for over a month but delayed due to difficulties in getting it reviewed and difficulties around agreeing on a memory model that'll make sense in a post oil-pan world. Work completed about ten days after the end of quarter.
Integrate font related memory consumption to memory-infra
Owner(s):drott Score:1.00
[mid quarter] On track. [end of quarter] Memory instrumentation added for key font metrics.
Have fullscreen implementation not modify layout tree
Owner(s):bokan Score:0.00
[mid quarter] De-staffed. At risk. [end of quarter] De-staffed as bokan wasn't available to layout-dev to work on fullscreen or layout in general. Needs new owner.
Have flexbox match latest version of spec
Owner(s):cbiesinger Score:0.80
[mid quarter] On track. [end of quarter] ?
Work with other vendors to nail down Custom Layout API
Owner(s):ikilpatrick Score:0.90
[mid quarter] On track. [end of quarter] Several meetings and discussions with other vendors., Have broad consensus and expect us to be able to spec it in Sydney.
Support Blink/Android hyphen design work
Owner(s):szager Score:1.00
[mid quarter] On track. [end of quarter] Sufficient support for design work provided. drott/szager up to speed.
Replace flexbox tests with W3C ones
Owner(s):cbiesinger Score:1.00
[mid quarter] On track. [end of quarter] Done?
Unprefix CSS Writing modes
Owner(s):kojii Score:0.70
[mid quarter] On track. [end of quarter]
CSS font-feature-settings
Owner(s):kojii Score:0.00
[mid quarter] At risk. [end of quarter] ?
Have root layer scrolling pass all unit tests
Owner(s):skobes, szager Score:0.50
[mid quarter] De-emphasized to focus on intersection observer. Still making solid progress though. [end of quarter] Some progress, but less than expected due to Stefan being pulled of to do Intersection Observer. Steve fixed at least one of the unit tests and a few of the layout tests, and added infrastructure for tracking future progress on the layout tests via flag-specific expectations.
Have prototype of ruby on top of custom layout [kojii, houdini-dev]
Owner(s):blink-houdini, kojii Score:1.00
[mid quarter] Rough prototype exists, needs further work. [end of quarter] Two independent prototypes created using slightly different APIs. Successfully demonstrated that it is viable to re-implement ruby on top of a custom layout API. Neither are production ready and both depend on a polyfil, as planned.
Remove simple text
Owner(s):eae, drott Score:0.90
[mid quarter] Always use complex text enabled on trunk. Plan is to remove code for M50. [end of quarter] Harder than anticipated due to a number of compatibility and perf issues, most severe being a perf regression in Mac AAT backend due to n squared font fallback which took awhile to resolve. Patch to remove simple text uploaded and approved. Will commit as soon as M49 has been branched. Missing end of Q4 goal by two weeks.
Ship improved font fallback to stable
Owner(s):drott Score:0.75
[mid quarter] Shipped to beta. On track. [end of quarter] In 48 which goes to stable in Jan 26. Missing end of Q4 goal by one month.
Finish wrapping block layout in an API
Owner(s):leviw, pilgrim Score:0.70
[mid quarter] At risk. [end of quarter] Still not quite done but getting closer.
Add class level documentation to all key layout classes
Owner(s):jchaffraix Score:1.00
[mid quarter update] On track. [end of quarter] Comprehensive class level documentation added as header comments to all key layout classes.
Feasibility study for replacing flipped flocks writing mode implementation
Owner(s):wkorman Score:0.00
https://crbug.com/535637 [mid quarter] Delayed to focus on paint performance. At risk. [end of quarter] De-emphasized to focus on slimming paint which is higher performance. Will carry over to Q1.
Continue to triage all bugs within one week
Owner(s):eae Score:1.00
[mid quarter] On track [end of quarter] All bugs triaged within one week of filing (for new bugs) or one week of having the CR-Blink-Layout label added (for old bugs, working down the backlog).
Fix all (new) P0/P1 within one release
Owner(s):eae Score:0.80
[mid quarter] On track [end of quarter] Still some ambiguity aroud the definition of a P1 bugs. All P0 and release-blockers fixed within one release. 75% of all P1s, majority of remaining ones should like have been classified as P2.
Reduce untriaged bug count by 50%
Owner(s):eae, jchaffraix Score:1.00
[mid quarter] On track to exceed goal. Reduced by 60% thus far. [end of quarter] Reduced by over 70%, far exceeding goal of 50%. Down to 180 unconfirmed/untriaged from 690 at the start of the quarter.
Reduce total bug backlog by 15%
Owner(s):eae Score:1.00
[mid quarter] On track to exceed goal. Reduced by 16% thus far. [end of quarter] Reduced total open bug count by over 25%, exceeding goal of 15%. Down to 1688 open bugs from 2310 at the start of the quarter.