4 ms·
On the unguessability of other addresses - I rotate several schemas making the mask of the address something like [service].[username].$(pwgen -1 | tr '[:upper:
by VTimofeenko 5y ago
On the unguessability of other addresses - I rotate several schemas making the mask of the address something like [service].[username].$(pwgen -1 | tr '[:upper:]' '[:lower:]')@customdomain.
Sometimes "[service]" is also shortened like "hackernews -> hn" to dodge the ban on service name in the e-mail address that some service providers apparently have.
- Vinnl 5y agoAh, I don't necessarily mean guessability of which other addresses I use, but of how you can reach me. If I block yourservice@mydomain.com, you can still attempt to reach me at totallynotyourservice@mydomain.com and it will work. You'll also be able to link my different addresses on different services. If I throw away the Relay alias for your service, that's it - there's no way to lead that back to me anymore.
- VTimofeenko 5y agoGotcha. > you can still attempt to reach me at totallynotyourservice@mydomain.com not if the catch-all address is actually /dev/null and the totallynotyourservice@ has to be mined from somewhere because it's random. Overall, I think it depends on the obfuscation strategy. It's true that having a unique @mydomain.com part is a big giveaway and someone could theoretically track one's activity by searching for all e-mail addresses coming from the domain. My use-case is more to use unique e-mail addresses to throw off credential stuffing attacks, not become untrackable/avoid all spam. For the tracking use-case I generally think several times if I want to register somewhere at all and try the usual routes first (mailinator, random old addresses on public e-mail).