Re: [802.21] Comments on 21-06-0698-01-0000_IP_configuration_methods_IE.doc
Hi Qiaobing,
As I mentioned below, if a proposal in 802.21 does not need
any modification to IETF protocols, there is no need for 802.21 WG to
communicate with IETF (i.e., no review by IETF is needed).
If a proposal defined in 802.21 needs modification to IETF protocols,
then a gap will be created if the proposal is accepted in 802.21 but
the actual modification to IETF protocols is rejected in IETF for some
reasons (e.g, interoperability issue, etc.). A similar situation is
happening in 3GPP2 regarding their EAP over HRPD proposal that relies
on IETF to define a shim layer between HRPD layer and EAP layer, which
is getting a strong push back from IETF. That kind of situation
should be avoided as much as possible, and proactive liaison procedure
is really desirable. If such a liaison procedure is not possible,
then the only way would be to make a decision without communicating
with IETF, but we must be aware of the potential risk of creating a
gap.
In general, the bar to accept a proposal that belongs to the latter
category should be higher than the former, IMHO (and the same argument
should apply to any 802.21 proposal that is closely related to
protocols defined in other WGs and SDOs.)
Of course this is my personal view and please correct me if my view is
wrong/biased/unreasonable.
BTW, I am still not sure whether the proposed IE is beneficial without
modifying IETF protocols or not. We can continue the discussion
either online or offline.
Regards,
Yoshihiro Ohba
On Tue, Aug 08, 2006 at 12:02:10PM -0500, Qiaobing Xie wrote:
> Hi, Yoshi,
>
> ...
> >>You are making these arguments under the assumption that "IETF will do
> >>nothing to support 802.21 and therefore we have to restrict ourselves
> >>within the current IETF protocol behaviors". This is not the right
> >>reasoning.
> >
> > If you have that impression in my argument, then it is not correct.
>
> Glad to hear that form you. At least we agree that 802.21 design is not
> restricted to the current upper layer solution.
>
> >
> > My standpoint is:
> >
> > - We should not define something new without understanding the benefit.
>
> Of cause.
>
> >
> > - If 802.21 define something new and beneficial and without need to
> > change IETF protocols, then we don't need to communicate with IETF for
> > it.
> >
> > - If 802.21 define something new and beneficial but that needs some
> > modification to IETF protocols, then we should communicate with IETF
> > so that IETF can analyze whether the modifications do not affect
> > interoperability and make actual modification to the IETF protocols
> > accordingly. I am happy to function as a communication channel
> > between 802.21 WG and IETF.
>
> I have no problem if you or anyone else want to talk with IETF folks
> about the technical details on how *in the future* a particular IETF
> protocol may be modified to take advantage of something from 802.21. But
> I have a BIG problem if you try to make IETF consent/ok a requirement
> for 802.21 to make its design decision. I hope I am wrong but this is
> the impression I got from your arguments so far.
>
> But if you take a look at the reference MIH models in section 5.5, IETF
> is hiding behind the "MIH Users (L3MP, Handover policy, Transport,
> Applications)" above MIH_SAP. It is just *one* of our many envisioned
> *potential* MIH users. We have in the draft spec a rich set of IEs
> containing information/parameter/attributes related to other
> technologies (IETF, 3GPP, 3GPP2, .16, .11) and we envision those IEs
> will be used by many many potential MIH Users. However, we have never
> required the potential MIH User to review and ok those IEs before we
> accepted them in the draft spec. Why should this particular IE proposal
> require the review and ok from a potential MIH User - IETF?
>
> If your reasoning stands, and as a fair-minded individual, it would make
> sense to you that all the IEs defined 802.21 would need to go through
> the same MIH User review/ok process. Do we want to do that? For example,
> for each of the 3GPP, 3GPP2, .16, .11, information we want to include in
> 802.21 IS, do we need to first communicate with 3GPP, 3GPP2, .16, .11
> and ask them to review and see if the information included in the 802.21
> IE would be misused by an MIH User and cause interoperability problem?
> Also, are you going to ask for "actual modification" how the potential
> MIH User will use each IE before you agree with it in 802.21?
>
> If the group agrees that should be the process, I will be glad to go
> along with it. But it would need to be applied to all the IEs.
> Otherwise, I would consider you are trying to apply a double standard to
> this particular proposal.
>
> regards,
> -Qiaobing
>
> >
> > Regards,
> > Yoshihiro Ohba
> >
> >
> >
> >>regards,
> >>-Qiaobing
> >>
> >>
> >>>Regards,
> >>>Yoshihiro Ohba
> >>
> >>
> >>
> >
>