Re: [netext] yet another charter version

["Koodli, Rajeev" <[e-mail]>]

Okay, but I thought he is fine with Jari's proposed text. Juan-Carlos?

-Rajeev
 

> -----Original Message-----
> From: Laganier, Julien [mailto:[e-mail]] 
> Sent: Thursday, January 28, 2010 5:53 PM
> To: Koodli, Rajeev; Zuniga, Juan Carlos; Jari Arkko; [e-mail]
> Subject: RE: [netext] yet another charter version
> 
> Hi Rajeev,
> 
> No disagreement. I understand that Juan Carlos had a concern 
> about excluding addressing single radio inter-tech handover 
> by using the term "simultaneously" to carachterize the 
> interface capabilities as in Jari's charter latest text. An 
> easy way to resolve his concern would be to add "and/or 
> sequentially" in the offending sentence.
> 
> OLD: It is assumed that that an IP layer interface can 
> simultaneously attach to multiple MAGs (possibly over multiple media).
> 
> NEW: It is assumed that that an IP layer interface can 
> simultaneously and/or sequentially attach to multiple MAGs 
> (possibly over multiple media).
> 
> Thanks.
> 
> --julien
> 
> > -----Original Message-----
> > From: Koodli, Rajeev [mailto:[e-mail]]
> > Sent: Thursday, January 28, 2010 4:41 PM
> > To: Laganier, Julien; Zuniga, Juan Carlos; Jari Arkko; 
> [e-mail]
> > Subject: RE: [netext] yet another charter version
> > 
> > 
> > Julien,
> > 
> > I don't see where we are disagreeing.
> > 
> > Jari's latest charter text is fine with me.
> > 
> > I do have an issue with an intermediate milestone (cf. my 
> other emails).
> > 
> > Otherwise, we are close to being done with this version.
> > 
> > -Rajeev
> > 
> > 
> > > -----Original Message-----
> > > From: Laganier, Julien [mailto:[e-mail]]
> > > Sent: Thursday, January 28, 2010 3:45 PM
> > > To: Koodli, Rajeev; Zuniga, Juan Carlos; Jari Arkko; 
> [e-mail]
> > > Subject: RE: [netext] yet another charter version
> > >
> > > Hi Rajeev,
> > >
> > > Koodli, Rajeev wrote:
> > > > [...]
> > > > > > 1. what does "sequentially or simultaneously attach"
> > > mean? What is
> > > > > > inter-access handover when you have simultaneous attachment?
> > > > > >
> > > > >
> > > > > There are two ways to perform a handover: 
> break-before-make, or 
> > > > > make-before-break. Both have advantages and 
> disadvantages at the 
> > > > > radio level that we don't need to discuss here. What we
> > > care is that
> > > > > in the break-before-make you can only have one radio 
> active at a 
> > > > > time and therefore you can only attach in sequence,
> > > whereas in the
> > > > > make-before-break case you can have two radios active at
> > > a time and
> > > > > therefore you 'may'
> > > > > attach to two MAGs simultaneously for a certain 
> period of time.
> > > > >
> > > >
> > > > Sure, there is break-before-make and make-before-break. How
> > > does that
> > > > come into play here?
> > > >
> > > > The problem we are trying to solve is how does the target MAG 
> > > > determine it is an inter-access handover, and the
> > > associated protocol extensions.
> > > >
> > > > We should be stating the problem rather than describe 
> radio level 
> > > > handover details, no?
> > >
> > > There has been no consensus for host changes to support network 
> > > based mobility management extensions for inter-access 
> handover and 
> > > flow mobility.
> > >
> > > So there's been a compromise in which we maintain the assumption 
> > > that the host is not changed and the interface between IP and the 
> > > network remains unchanged, and under this assumption, the 
> WG would 
> > > be allowed to develop inter-access handover and flow mobility 
> > > extensions.
> > >
> > > The text I have discussed with Juan Carlos outlines the 
> assumption 
> > > under which the WG has to work to solve the problem.
> > >
> > > --julien
> > >
> > >
>