On 2008-11-18 10:35, Daniel Park wrote:
> Confusing...definitely RFC5121 must be implemented for IPv6 over WiMAX
> network. EthernetCS must be too. No need to decide which CS must be used for
> WiMAX by 6man...?
My understanding is that in the IETF we try to avoid situations where
two ways of doing the same thing MUST be implemented. I hope that 16ng will
produce an applicability statement, but I don't see what we can usefully
say in the node requirements.
Brian
>
> Daniel
>
> On Tue, Nov 18, 2008 at 5:20 AM, Brian E Carpenter <
> [EMAIL PROTECTED]> wrote:
>
>> On 2008-11-18 09:03, Daniel Park wrote:
>>> Brian (ccing 16ng list)
>>>
>>> On Mon, Nov 17, 2008 at 12:12 PM, Brian E Carpenter <
>>> [EMAIL PROTECTED]> wrote:
>>>
>>>> Daniel,
>>>>
>>>> On 2008-11-14 22:19, Daniel Park wrote:
>>>>> John,
>>>>>
>>>>> I would make a minor change according to the valuable comments from the
>>>>> WiMAX expert:
>>>>>
>>>>> There are two ways to transfer IPv6 over WiMAX:
>>>>> - IPv6 over WiMAX using IPCS: RFC5121
>>>>> - IPv6 over Ethernet carried over WiMAX: AD Evaluation (ID Status)
>>>> Where does your suggested preference for the IPv6 CS come from?
>>>> Does 16ng have a consensus on this preference?
>>>>
>>>> Since I can't imagine anyone *not* implementing the Ethernet CS,
>>>> doesn't this make extra work for all implementors, compared
>>>> with preferring the Ethernet CS for both IP versions?
>>>
>>> We had long discussion on that before..:-) Yes, IPv6CS (RFC5121) comes
>> from
>>> 16ng WG. And EthernetCS is too. Prebably, the consensus you mentioned
>> above
>>> means which CS is a mandatory or optional for IPv6 implementation or
>>> both. Well, 16ng WG just leaves them to the business choice in WiMAX
>>> networks since that is beyond scope of 16ng. As of today, obviously
>> mobile
>>> biz wants to implement IPv6CS only, and wired biz (looks like DSL) wants
>> to
>>> implemt EthernetCS. But, no one knows what happens tomorrow in WiMAX
>>> networks.
>> Then it seems to me there is no useful statement to be made
>> in the node requirements document. If 16ng has no recommendation,
>> I don't see how 6man can decide.
>>
>> Thanks
>>
>> Brian
>>
>
--------------------------------------------------------------------
IETF IPv6 working group mailing list
ipv6@ietf.org
Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6
--------------------------------------------------------------------
