You know, I’m generally a privacy conscious consumer and default to opting out of analytics, but if a company asked me to do a survey about how I use the product I absolutely would! I enjoy giving feedback, but am always going to pass on dragnet surveillance.
I’ve never seen this though. Perhaps I’m naive to certain business interests or statistical techniques, but it seems like such a survey would be a necessary part of a robust decision making process based on mass-surveillance gathered analytics for a product that doesn’t earn money by directly selling that data. (if you only make money by tracking users, I understand why you might not care about users who prevent it, but even then with the power of network effects I’d expect a business would at least want a finger on the pulse)
I mean, I don’t have a degree in stats, but from what I do know this is the kind of due diligence that the foundation of trust in statistics is based on. Maybe I’m just confused and someone can correct me, but I’ve come to see the selective enforcement of good statistical practices as one of the major problems in the world today, because a) I see it so much, b) it gives certainty where little is deserved, c) it prevents us from getting to truth.
It's a bit worse than even this suggests. Using any feature is correlated with having analytics disabled, because the more of a power user someone is, the more they'll use "advanced" features (anything beyond "clicking in the URL bar and typing a search term"), and obviously power users are the only people disabling analytics.
So if anything there's an all-encompassing metrics bias saying "users want fewer features and less complexity".
There is no maintenance cost as they are keeping the "touch" option. Unless checkboxes are somehow easier maintain than drop-downs.
It's about cost of maintenance not people wanting a feature to be removed. And while this feature specifically doesn't have to high of a cost (I think) it will sum up if many "hardly used" features are left in place.
> >The Engineering team is well aware of the community > feedback. > > Does that mean Mozilla is aware how unpopular this is, but are going to go ahead nevertheless?
I think it means they decided it, then got feedback and now are potentially reevaluating it.
Also maybe the group of people using it isn't very high but very vocal/loud, in which case going ahead anyway would be reasonable.
Through I also would guess the group of people using that feature is also likely to have analytics disabled...
Having "analytics disabled" correlate with "using that specific feature" is always a major problem for data driven decisions, worse it's a well hidden and hard to estimate problem.