Re: [sip-overload] Local Policy Re: I-D Action: draft-ietf-soc-load-control-event-package-05.txt

["Vijay K. Gurbani" <[e-mail]>]

On 01/14/2013 11:27 AM, DRAGE, Keith (Keith) wrote:
> I'm not convinced we have got to the bottom of the requirement yet.
> Writing SHOULD is not an excuse for writing a poor requirement, yet
> that seems currently to be the excuse for wring SHOULD instead of
> MUST. Ideally I should still be able to still identify conformance or
> no conformance to a SHOULD requirement.
>
> I suspect what we need to do is find a more specific requirement and
> then cover the rest of the problem in more informative text, rather
> than trying to address everything in a single requirement sentence.
>
> I agree that if we end up writing a SHOULD, then there should be
> associated informative text that identifies the example circumstances
> for ignoring the SHOULD. That has long been an unwritten rule, and
> without it the SHOULD always gets interpreted as a MAY.

Keith: Do you have specific text you'd like to suggest?  Taking in
account Sal's view of staying away from local policies in conflict
and taking Janet's and Charles' suggestions, I have the following as a
cautionary paragraph before the normative text:

    It is expected that a SIP client will follow local policy as long as
    the results in reduction of traffic is consistent with the SIP
    overload algorithm in effect at that node.  Accordingly, the
    normative behaviour in the next three paragraphs should be
    interpreted with the understanding that the SIP client will aim to
    preserve local policy to the fullest extent possible.

And then we go down the three SHOULD paragraphs.

I believe that suggesting specific text will help close this gap more
expeditiously.

Is the above something that we can go with?

Thanks,

- vijay
-- 
Vijay K. Gurbani, Bell Laboratories, Alcatel-Lucent
1960 Lucent Lane, Rm. 9C-533, Naperville, Illinois 60563 (USA)
Email: vkg@{bell-labs.com,acm.org} / [e-mail]
Web:   http://ect.bell-labs.com/who/vkg/