Re: [Asrg] rDNS

[Alessandro Vesely <[e-mail]>]

der Mouse wrote:
>> IMHO, synthesizing informations on that basis is symptomatic of
>> technologies in their infancy being desperately greedy for data they
>> cannot obtain [otherwise].
> 
> "Synthesizing" information?  What's synthetic about (paraphrased) "your
> rDNS is missing or broken, so I won't listen to you"?

One of the implied major premises is synthetic. The complete logic 
chain reads:

* I won't listen to clueless hosts (arbitrary, but legitimate premise)
* all hosts with a missing or broken rDNS are clueless (synthetic)
* your rDNS is missing or broken (fact)
* so I won't listen to you (conclusion)

If I lived in a land where all ISPs obey a policy of not setting rDNS, 
I could opt for setting up a host in a different country, give up 
rDNS, or emigrate. The second choice if not bad, given that I like to 
see the hardware and have a few reasons to tread this soil. Local ISP 
market should not affect the reliability of hosts operating in that area.

> [rDNS is required] because it correlates with having clue 
> (thereby being less likely to be a problem).

I'm not holding that it's wrong. Possibly, a good deal of ISPs do 
rfc2317 delegations (if the customers insist.) Hence, statistically, 
those who don't set up rDNS are clueless. However, general assumptions 
based on statistical indications are synthetic. The fact that you have 
to use them indicates that email technology (or ISP market) is immature.

Specifically, the "iprev" authentication method described in rfc5451 
should not be strictly required. At least, if an SMTP connection is 
not accepted (or accepted with high suspicion) because of that reason, 
the client should be made aware of this circumstance, possibly 
allowing it to suggest alternative authentication methods or relay its 
mail through alternative paths. (The second part of this paragraph is 
the subject of draft-vesely-vhlo.)