Re: [apps-discuss] Review of draft-melnikov-smtp-priority-04

[Alexey Melnikov <[e-mail]>]

Hi Glenn,
To answer your question about the number of priority levels:

On 10/01/2012 21:10, Glenn Parsons wrote:
> I have been selected as the Applications Area Directorate reviewer for 
> this draft (for background on appsdir, please see 
> _http://trac.tools.ietf.org/area/app/trac/wiki/ApplicationsAreaDirectorate_). 
>
> Please resolve these comments along with any other Last Call comments 
> you may receive. Please wait for direction from your document shepherd 
> or AD before posting a new version of the draft.
> Document: draft-melnikov-smtp-priority-04
> Title: Simple Mail Transfer Protocol extension for Message Priorities
> Reviewer: Glenn Parsons
> Review Date:  January 10, 2012
> Summary:
> This draft is not ready for publication as a Proposed Standard and 
> should be revised before publication
> Major Issues:
[...]
> 5. The priority level definition is not clear.  There are 200 possible 
> values.
The document originally had 5 fixed levels. I received comments from 
some people that they would like to use this extension, but for their 
environment they wanted to use N levels (N > 5), or maybe M levels, but 
they were not quite the same as the one recommended in the draft. One 
proposal asked for having a language for expressing policy associated 
with each priority level and use names instead of numbers. I felt that 
use of names for priority levels was overkill, so instead I've expanded 
the range of allowed numeric priorities. The main nice property of this 
is that the range of allowed priorities allows an implementation to 
insert additional priority levels in gaps between recommended priorities 
(e.g. if -8, -6,-2,0,+2,+6,+10 are defined, one can define a new 
priority level of +7 if really needed).
> But there MUST be at least 6 supported.  OK that seems like overkill 
> -- why can't it be from -9 to 9 then?
I think this will work. I was certainly not thinking that all 200 
distinct priority levels should be implemented.
> Anyway, is that 6 distinct numbers, 6 distinct numbers from the range 
> shown, or any number for the range shown?
Any number from the range shown, but at least 6 of them.
> Further I would  suggest that a default value should be specified for 
> these 6 levels and the default mapping ranges -- all in a table.
**Ok.

So it looks like I need to (a) reduce the number of allowed priorities, 
(b) clarified how they are to be used and why there are more than 6 and 
(c) recommend a default mapping.