Benjamin Smedberg Reporter
	Description • 14 years ago

I want to be able to measure whether users choose to submit plugin crash reports. The basic data is:

* when the plugin crashes, we check whether the UI code decided to display the in-page UI or the infobar UI for each tab. We record how many tabs are affected and the UI shown, and which tab was in the foreground for each window

* when the user chooses to submit, we record which tab they chose to submit in

It's not clear to me whether this is better as a testpilot study or a telemetry thing.

	(dormant account)
	Comment 1 • 14 years ago

Currently test pilot has a bigger enabled-shared than test-pilot. So for transitional stuff like this, the best way forward would be to add a telemetry probe and then also report it via test-pilot. This way as our telemetry share eclipses test-pilot the stats don't need to be reimplemented.

	Jinghua zhang
	Comment 2 • 14 years ago

Adding Jono to this bug and let's figure out what we should do as the next step.

	Justin Dolske [:Dolske]
	Comment 3 • 13 years ago

(In reply to comment #0)

> * [...] We record how many tabs are
> affected and the UI shown, and which tab was in the foreground for each
> window

What do we hope to learn from this? (Ditto for recording the tab the crash was reported from.) Just curious.

Could this just be two counters? One for the number of crashes, one for the number of reports submitted?

	Benjamin Smedberg Reporter
	Comment 4 • 13 years ago

I'd like to learn:

* whether users typically submit the crash report immediately from the page they are looking at when the crash occurs, or whether they may submit it "later" from another tab. I'd also like to get a sense for the proportion of in-page UI versus infobar UI.