ietf-dkim
[Top] [All Lists]

Re: [ietf-dkim] Alternative MAiling List Approach

2010-08-04 09:58:23
"1. Only modify the From: in cases where the incoming mail was 
DKIM-signed."

Like I said, this screws up the mail for the 99% of people who don't use 
ADSP. since they can no longer respond directly to the authors of (some) 
list mail.

No it doesn't. An essential part of the scheme was that the MLM should 
provide a mechanism so that replies sent to the mutilated From: would get 
delivered correctly ...

Sigh.  They're not responding directly, they're responding via a kludge 
run by the MLM.  As an example of things this breaks, I send a lot of mail 
to people I know in response to mail they sent through a list.  (This 
message, for example.)  The way my mail is filed, I can search by address 
to see what I've written to whom.  Your hack breaks it.

Many MUAs make addressbook entries for addresses in outgoing mail. 
Should it do that for these hack addresses?  If the person unsubscribes, 
should the forward still work?  If not, you've just broken address books. 
If so, you've just reinvented an open relay.

(No, I have no interest in changing my mail software to work around your 
damage.)

As I also noted, people who believe that they are so important that it's 
essential for recipients to verify that mail is really from them can use 
S/MIME, which is if anything more likely than DKIM to survive list 
mangling.

That might be a fine thing, but you then have the problem of educating a 
large number of people to do something they do not currently do (and likely 
don't even know how to do).

I can't help but note that to about five significant digits, nobody is 
sending personal mail and marking it discardable.  Isn't the best response 
to a small number of people who want to do something silly is "don't do 
that"?

R's,
John
_______________________________________________
NOTE WELL: This list operates according to 
http://mipassoc.org/dkim/ietf-list-rules.html

<Prev in Thread] Current Thread [Next in Thread>