Re: Last Call: draft-klensin-rfc2821bis

Ned Freed wrote:
>> Indeed, if you asked
>> a random sampling of those groups --remembering that there are a
>> huge number of SMTP servers in the world, only a tiny fraction
>> of which are professional operations and with an even smaller
>> fraction being large-scale, carefully-managed production ones,
>> you might discover that many of them had forgotten that there
>> was such a thing as an MX record and how to set it up.
> 
> And even if they know MX records exist they may not be able to use them. Some
> DNS provisioning arrangements allow users to set up MX records but there are
> others that do not.

This observation moves the proposed AAAA-only mode into one of high risk. 
Rather than the benefit of reducing complexity, it becomes the danger of 
threatening interoperability.

Let's note that the proposal is a change from a model that has worked solidly 
for 20 years and has not been a source of problems.  While the proposal is for a 
mode that is appealing, it is not necessary.

Methinks it is therefore time to respectfully retire the proposal.

d/

-- 

   Dave Crocker
   Brandenburg InternetWorking
   bbiw.net
_______________________________________________
IETF mailing list
IETF@ietf.org
https://www.ietf.org/mailman/listinfo/ietf