Quote:
Originally Posted by JeremyNicoll
Nonsensical - because there's no way that FM could verify it so they shouldn't be trying to. (They should probably also disallow wildcard identities for target domains that are not entirely owned by the user.)
|
I wrote the "They should probably also disallow..." part before I read how xyzzy uses eg *@service.letterboxes.org so clearly they'd need to allow that sort of thing.
It also occurs to me that if eg xyzzy subsequently deleted their registration of
service@letterboxes.org FM's code would have to go looking for any matching wildcard definitions and delete them too. I wonder if FM's code does that now?
It seems to me that the whole wildcard thing is more complicated than it needs to be. Wouldn't it just be simpler if -
- for personal domains, hosted by FM so they know who owns it, the Compose screen allowed the owner to edit anything to the left of the "@" in a sender's address
- for personal domains, hosted elsewhere, but entirely owned by the customer [who'd have to prove that somehow], the same
- for FM-owned domains, the Compose screen allowed a user to edit anything to the left of the entire registered email address. That'd allow xyzzy to enter eg "grocery" and an implied dot to the left of "services@letterboxes.org" to form "grocery.services@letterboxes.org", and it would allow me to change my apparent name from "Jeremy Nicoll" to "Some kind of idiot"
- for individual addresses at domains hosted elsewhere, each address verified as is being done now, the Compose screen allowed a user to edit anything to the left of the entire verified email address.
You wouldn't need wildcard definitions at all, just more sophisticated logic in the Compose screen.
And another thing... Just because I can verify today that I have access to emails sent to a particular address doesn't mean that that will be true for evermore. How often will FM re-verify these addresses?