On 26 mar 2015, at 19:57, Doug Montgomery <[email protected]> wrote:
> I guess I am confused as to why I would want to support either key discovery
> or distribution for the (500million - 1) versions of my email address that I
> don't use.
>
> My mail agent only generates one version of my from address. I only convey
> one version in written, oral, and online forms of distribution.
>
> I am perfectly happy if it is not possible to discover the 500million-1
> variants of my email address that I don't use or exercise but that my mail
> provider chooses to treat as equivalent. Frankly if it was possible, I
> would write mail filters to drop those variants on receipt, because they
> clearly indicate some use of my email address that I did not endorse.
>
> For my uses, straight hashing of the form of my address you are presented
> with is all I want. To be honest, I would consider it an undesirable feature
> to support the versions of my address that I don't use.
>
> Enabling the simple direct function of key retrieval for the addresses we
> use, in a ubiquitous way would be very powerful.
>
> We should leave for future study the issue of key retrieval for the addresses
> we don't use.
I'm completely agree with Doug here. MTAs may do all kinds of magic matching,
what matters is what address the sender/receiver use and providing a solution
for that is very powerful.
jakob
_______________________________________________
dane mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/dane
