[ietf-dkim] Policy vs Practices

[Douglas Otis]

On Sep 5, 2006, at 8:48 AM, Hallam-Baker, Phillip wrote:

>
> Using the term 'practices' instead of policy is not going to fool  
> anyone, the objection is to the idea, not the name. The term  
> practices is highly loaded and has its own meaning and will cause  
> even more people to shudder.

The term "practice" over "policy" acknowledges information conveyed  
is not action related instruction for the verifier.  This information  
instead offers assurances regarding an initial state of the message.   
While DKIM is a good mechanism for safe message annotation (provided  
other assurances are made), signature requirements for acceptance  
will disrupt many common email services.  Within the current DKIM  
limitations, rigid verifier instructions would not be compatible with  
email in general.  The term "practice" better conveys this design  
limitation, and better clarifies how this mechanism might be used.

-Doug