On Jun 11, 2007, at 9:00 AM, Steve Atkins wrote:
On Jun 11, 2007, at 8:55 AM, Douglas Otis wrote:
On Jun 11, 2007, at 5:43 AM, J.D. Falk wrote:
On 2007-06-09 17:01, Hector Santos agreed with John Levine:
"It's all spam" is about the simplest useful advice a (non)
sender can give. In my case, which I don't think is unusual, I
get buckets of spam and blowback to subdomains that have never,
ever, sent a real message. The domains are the names of
computers on my network, which were probably scraped out of
usenet or mail archive message IDs. If receivers were to
reject or drop all mail purporting to be from those domains, it
would be uniformly better both for the receivers (less spam,
cheap filter) and for me (less blowback.)
+1
+1
'Proof of use' as acceptance criteria should also eliminate much
of this. Deprecating and then obsoleting A record discovery would
safely arrive at this goal.
Use of * MX "." would be something a root server will be unable to
defending against. If this were to become popular, these records
may lead to a type of DDoS attack against the root servers. This
construct assumes millions of email applications will be modified
to recognize "." as an invalid domain name.
Why do you believe this?
Largely based upon the advice of a root operator. May I suggest this
issue be taken to a DNS WG. Few protocols compare to email's ability
to directly induce third-parties into making DNS transactions. As
such, email design should be cognizant of dangers associated with
this anonymous push model.
-Doug
_______________________________________________
NOTE WELL: This list operates according to
http://mipassoc.org/dkim/ietf-list-rules.html