Closed Bug 1365432 Opened 3 years ago Closed 3 years ago

Conduct A/B study on HSTS priming

Categories

(Core :: DOM: Security, enhancement, P1)

55 Branch
enhancement

Tracking

()

RESOLVED INVALID

People

(Reporter: kmckinley, Assigned: kmckinley)

References

Details

(Whiteboard: [domsecurity-active])

Collect data on HSTS priming vs no HSTS priming in the wild, including the following:
* Gather better data on how many requests are mixed-content and HSTS priming
* Comparative metrics - do HSTS priming-enabled users see more HSTS upgrades?
* Evaluate the impact of HSTS priming on load times
* Gather data on HSTS upgrades by source (preload, natural, HSTS priming)

This bug is to create an addon to divide users into up to three groups, a group where the HSTS priming preferences remain off (control), a group where HSTS priming preferences are flipped on, and optionally a group where the pref to send HSTS priming requests is off and the pref to swap HSTS and mixed-content blocking is on. The third group would help to understand the impact of the HSTS priming request itself.

We should be able to use a Shield study to control the addon[1]. This would give us an easy way to roll out the study to a small number of Release users, as well as neatly gathering the resulting telemetry for analysis.

This study is targeted for 55 Beta.

[1] https://github.com/gregglind/shield-study-embedded-web-extension-example/blob/master/stage-2-shield-study/addon/lib/ShieldStudy.jsm#L141-L148
Blocks: 1246537
Priority: -- → P1
Whiteboard: [domsecurity-active]
It has been determined that the work is more than needed for this.
Status: ASSIGNED → RESOLVED
Closed: 3 years ago
Resolution: --- → INVALID
You need to log in before you can comment on or make changes to this bug.