[Netext] next steps for netext

[julien.laganier.ietf at googlemail.com (Julien Laganier)]

Hello Sri,

On Thu, Apr 9, 2009 at 12:07 AM, Sri Gundavelli <sgundave at cisco.com> wrote:
>
> [...]
>
>>> So, its possible for the MAG to derive that information from the
>>> network, in many cases, including in 3GPP, as Vijay pointed out
>>> for single attach.
>>
>> This is done by a special L2 message introduced for that - i.e. a change
>> of the host (which is not visible to IETF as it is done at L2 and not L3,
>> but still very real)
>
> We can indeed define a L2 message, allow the network to make that
> decision, or allow it to push policy templates to the terminal.
>
> In any case, changes to host, in the form of application requirement
> or connection manager requirement, or any application layer policy
> push, does not break IETF standards. If some operator wants to install
> such application on the terminal, I wonder why we should have a problem.
> Also, IMO, we are dreaming, if we assume that any operator will qualify
> any host on their network, with out zero software addition. Looking
> at EV-DO deployments today, all operators ship connection manager
> software.

I don't know about EV-DO -- But for sure I wasn't dreaming the many
times I accessed an HSPA network with my off-the-shelf Linux laptop
(i.e. _with_ zero software addition.)

--julien