[ietf-dkim] draft Errata on RFC 4871

[Dave CROCKER]

Eliot Lear wrote:
> I think perhaps it would help, Dave, if you could step through the 
> ramifications of your concerns. 

I don't understand your question.  What is it about the Introduction to the 
Errata that is not sufficiently clear or complete?

You appear to be asking about the ramifications of being non-interoperable, and 
I know you don't really mean that.


>    What it doesn't say is how the confusion 
> manifests itself.  Can you please clearly do so with examples?  

You need an example of the problems that can ensue from having one side of a 
protocol exchange consider one attribute to be the sole output and another side 
consider two attributes or a different one attribute to be the output?

Really?


>> If the UAID is not the same as the address in the From:
>>       header field, the mail system SHOULD take pains to ensure that the
>>       actual UAID is clear to the reader.
> 
> If you are going to go to lengths to separate the Identity Assessor from 
> other parts of the system, it makes it more confusing when you continue 
> the old phrasing of "mail system" above.  Which mail system?  Who is 
> responsible for that?  I presume you mean the signer, since the signer 
> is the one who inserts the UAID.

damn. damn. damn.  yes it does make it more confusing.  yes I meant to change 
it.  grrrr.  sorry.

I believe the intent of that sentence is something like "receiving MUA", but on 
reflection, I'm not sure I fully understand what this requirement really is or 
what is considered reasonable for satisfying it.  And I mean that as a 
specification-clarity issue, not an user interface philosophy issue.

d/
-- 

   Dave Crocker
   Brandenburg InternetWorking
   bbiw.net