Re: [Dime] Ben's comments on 5.2.2: 5th paragraph

On 1/15/15 9:15 AM, Ben Campbell wrote:
>> On Jan 15, 2015, at 7:38 AM, Steve Donovan <srdonovan@usdonovans.com> wrote:
>>
>>
>> On 1/15/15 3:59 AM, Wiehe, Ulrich (NSN - DE/Munich) wrote:
>>> Ben wrote:
>>> -- 5.2.2: 5th paragraph:
>>>   
>>> This doesn't seem quite right, since it leaves the option of no abatement at all. Second, it doesn't seem to allow delegation of abatement downstream. When might one choose to ignore those SHOULDs? Is this here to allow delegation? If so, the use of SHOULD makes local throttling preferred over delegation.
>>>   
>>> <Ulrich> My understanding is: If delegation of abatement is done by a node, then that node is no longer a reacting node.
>>> The first SHOULD is ignored by reacting nodes that do not support diversion and hence always perform throttling. For the second SHOULD you MAY be right. We SHOULD replace it with MUST.
>>> In addition the world “otherwise” must not be read as “ if diversion abatement treatment is not possible” but as “ if diversion abatement treatment is not possible and if the first SHOULD is ignored”.
>> SRD>  I mostly agree with Ulrich.  I propose the second sentence be changed to:
>> The reacting node MUST apply throttling abatement treatment to requests
>> identified for abatement treatment  when diversion treatment
>> is not possible or was not applied.
>>
> I agree in general. But I suggest casting in the form of MUST apply an abatement treatment, which SHOULD be diversion if possible. That is, avoid baking the current fact that we have only two treatment possibilities into the normative language.  In the odd chance we come up with some new kind of treatment in the future, this form would require less modification.
SRD> Ok. I'll propose wording later.
>
>>>   
>>>
>>>
>>> _______________________________________________
>>> DiME mailing list
>>>
>>> DiME@ietf.org
>>> https://www.ietf.org/mailman/listinfo/dime
>> _______________________________________________
>> DiME mailing list
>> DiME@ietf.org
>> https://www.ietf.org/mailman/listinfo/dime
>