Specifically, 80% of profiles on release are less than 10 minutes a WEEK, according to John Jensen. Based on sampling and probability.
Bill On Wed, Jan 28, 2015 at 2:47 PM, Christopher Karlof <[email protected]> wrote: > On Wed, Jan 28, 2015 at 2:17 PM, Richard Newman <[email protected]> > wrote: > >> >>> - One less password: the password you use to sign into FxA, and to >>> unlock the vault, is one and the same. >>> - Recoverable: should you ever forget your vault password, you can >>> recover it via email (on your Mac or 1Password, if you forget your Master >>> Password, you’re hosed) >>> - >>> >>> It might be worth clarifying here that you can't actually recover the >> password from FxA. You can recover access to the same *kA* by resetting >> your password, at the cost of losing any data that was encrypted with kB >> (e.g., the contents of your Sync account) and re-entering your password on >> each of your devices. >> >> Avoiding these kinds of pitfalls are why key escrow was suggested earlier >> in this thread. It's really hard to be clear about exactly what these >> things mean, so maybe you meant something else — if so, please do clarify! >> >> Recoverability means either using kA, or using some stable value that's >> wrapped with kA and stored somewhere. >> >> >>> Of course, you can also say “My vault don’t need a password to unlock >>> it”. I suspect that most people who uses their own machine don’t want to >>> bother entering a password in order for Firefox to autofill. >>> >> >> Somewhere between 98% and 95% of Firefox users, last I checked, yeah. >> >> > We need to look at these percentages in release and correlate them with > Firefox usage hours. I really only care about these percentages in heavy, > medium users, and light users. I don’t care about lapsed users, i.e., < 10 > mins a day. These are a large chunk of users and I don’t want them skewing > data. > >
_______________________________________________ Dev-fxacct mailing list [email protected] https://mail.mozilla.org/listinfo/dev-fxacct
