> Hi Charlie,
> (2011/11/11 6:42), Charles E. Perkins wrote:
>> Hello Hidetoshi,
>> On 11/10/2011 4:41 AM, Hidetoshi Yokota wrote:
>>> I think that the point is to give a choice to the network and user if
>>> mobility is needed or not. Some applications benefit from seamless
>>> mobility (e.g., VoIP), but others may not (e.g., Web access). The
>>> network and user should be able to specify the capability of mobility
>>> and appropriate access network for each application. In that sense,
>>> Dynamic MM is an interesting feature.
>> I pretty much agree with this; I expect that working out the
>> details will be pretty tedious and perhaps even contentious.
>>> I agree that the current 3GPP architecture is very complicated and may
>>> not be optimal, but I don't think it is a good idea to change it in
>> I agree we can't "change" 3GPP architecture in the IETF, but
>> we can respond to the needs which are evident from considering
>> the 3GPP architecture. It seems very few believe that LTE
>> is really immune from further evolution and even significant
>> change in the future. If IETF does NOT respond to the needs made
>> evident in current 3GPP specifications, we have little hope of
>> proper integration into their future evolved designs.
> I fully agree with your point and am glad you initiated this discussion.
> One of the reasons for its complicatedness is that the 3GPP architecture
> has been incrementally built considering all the backward compatibility
> and interoperability along with introducing new features and minimizing
> the unhappiness of all. I respect this effort and achievement we all
> enjoy today. On the other hand, the current mobile traffic is way too
> much for this architecture to be able to handle. If IETF can contribute
> to tackling this issue toward the future evolved design, it will be very
> valuable. I don't know if DMM is the start point or we need a new WG,
> but we still have a high mountain to climb.
I fully aggree with this point. The mobile operators faces the presure
of rapidly traffic increasing. We should consider this point when we
try to optimize mobility protocols.