I’m assuming it’s a representative sample. Although self selected users may be more geeky... users have no control so it’s likely good.
Yes, at some point there will be a future version that hits ~50% and then it generally starts moving quicker.
Yeah I think the only way it would not be representative would be any active attempts TeslaFi users make at coaxing their SC to push updates or making frivolous service appointments.
And that might happen a little bit, but I highly doubt it's enough to substantially skew the dataset.