User:KaiRo/Goals

Personal goals of KaiRo per quarter (in addition to ongoing routine work).

longer-term ideas

Get graphs for categories of crashes (GC, OOM, shutdownhang, GFX, JIT, ...) to monitor if the volume of some category improves of gets worse.
Get a crash rate graph done that is useful for putting up on office screens

[ON TRACK] Create a script or dashboard using Telemetry data.
[ON TRACK] Start migrating the custom scripts that now are hg-hosted php using PostgreSQL over to git-hosted Python using Socorro API: Get at least one script into a github repository, using python and possibly the API instead of direct DB access.
[DONE] Improve Top Crash Score report to deal better with the Socorro API rate limit as well as to accept more parameters via URL and add drop-downs for selecting those. [implemented except the dorpdowns]
[ON TRACK] Reduce the amount of manual work done for routine tasks in Release Quality/Stability.
[ON TRACK] Continue learning more about people management at Mozilla (praxis insights on hiring, tools and structure used at Mozilla for people management). [being involved in interviews was a good step, more next quarters on this ongoing goal]

[DONE] Move arewestableyet.com to use the same numbers as the graphs and link graphs from there.
[DONE] Investigate Unified FHR/Telemetry as a source for better stability information. [partially done, Mozlando: sessions on Telemetry in general, bsmedberg's dashboard work]
[DONE] Train Release Management team to help in stability tasks. [partially done, potentially another round in Mozlando]
[MISSED] Start migrating the custom scripts that now are hg-hosted php using PostgreSQL over to git-hosted Python using Socorro API: Get at least one script into a github repository, using python and possibly the API instead of direct DB access.
[MISSED] Improve Top Crash Score report to deal better with the Socorro API rate limit as well as to accept more parameters via URL and add drop-downs for selecting those.
[DONE] Continue learning more about people management at Mozilla (praxis insights on hiring, tools and structure used at Mozilla for people management). [being involved in interviews was a good step, more next quarters on this ongoing goal]

[DONE] Tweak Top Crash Score to make it easier to use (e.g. color-code reasons for increasing/decreasing score). Note: additional pieces carried over to Q4.
[DONE] Get graphs up for per-channel volume of startup crashes, OOMs, shutdown hangs and un-symbolized signatures,
- [MISSED] and work on getting this data into metrics systems that can be used for executive dashboards.
[DONE] Get a prototype up for crash rate display on office screens.
[MISSED] Start migrating the custom scripts that now are hg-hosted php using PostgreSQL over to git-hosted Python using Socorro API: Get at least one script into a github repository, using python and possibly the API instead of direct DB access.
[MISSED] Investigate which areas of quality are unowned in Firefox/Platform at this point.
[AT RISK] Continue learning more about people management at Mozilla (praxis insights on hiring, tools and structure used at Mozilla for people management).

[DONE] Improve the "crash score" (relevance-weighted topcrash) prototype into a state that can experimentally be used for prioritizing crashes in release decisions, figure out how well it works for that and adjust factors if/as necessary.
[DONE] Highlight aspects of crashes that influence importance on the "crash score" prototype: startup, OOM, shutdown, number of installations, GC.
[MISSED] (Stretch goal:) Get graphs up for per-channel volume of startup crashes.
[DONE] Teach the full release management/quality team to run mozmill-ci update tests to remove myself from being a SPF while those still need to be run.
[DONE] Learn more about people management at Mozilla (at least hiring process, additional areas as possible).

[DONE] Create a concept and prototype for a better topcrash metric that (de)prioritizes for more/less serious issues.
[DONE] Familiarize with the current QE release process (more in depth) and build up some more routine about dealing with it.
[DONE] Find how we can make the QE release process more effective and refine the release test plan wiki pages.

[DONE] Learn more about how QA train driving works and what can be improved in that process.
[ON TRACK] Help rest of the Firefox QA team figure out how to focus work better, figure out how we could get metrics on that.
[DONE] Attend TRIBE: AoS.

[MISSED] Get up metrics for OOM and GC crashes in addition to the overall rates, figure out possible additional groups to measure (GFX? JS? JIT?).
[DONE] Automate more ongoing Firefox QA metrics, get a first version of a dashboard up.
[DONE] Get verification process into a more uniform and transparent state (esp. in terms of bug queries). [Queries on dashboard are more visible now, new flag exists and is used, but we are de-emphasizing verification overall.]
[DONE] Connect more with people on the QA team.

[DONE] Figure out if and how we can get metrics for certain categories of crashes (OOM? JIT? GFX?) in addition to the overall rates.
[DONE] Determine if we'll do a stability work week in 2014, and if so, start organizing.
[ON TRACK] Get more into assembling metrics for desktop QA.
[AT RISK] Get verification process into a more uniform and transparent state (esp. in terms of bug queries).
[ON TRACK] Learn more about QA-driving desktop trains, potentially start leading one.
[DONE] Figure out if PM is still needed for FHR, keep track of ongoing initiatives.
[DONE] For personal growth, get at least subscribed to the first module of TRIBE.

[AT RISK] Figure out project management needs in QA and start working on them.
[DONE] Extend analysis on and blog about crash rate histories, possibly also take a look into graphing certain categories of crashes (OOM? JIT? GFX?).
[DONE] Make sure that Data Quality, Stability Probes and User Tips initiatives on FHR move forward.
[AT RISK] Get an initial round-up on who would need to be involved in a Firefox OS FHR-or-alike and where those people/groups stand in terms of opinions on it.
[ON TRACK] To extend PM knowledge, learn about how scrum and similar processes are being used at Mozilla and also how hiring processes work here.