Re: RE: AD request / L2 Triggers Chapter Statement

Hi Behcet,

   I have not had time to read your charter.

   However, when I lead the pre-BOF meeting in Minn
   we used the above proposed charter.

Carl


----- Original Message -----
From: "Behcet Sarikaya" <[email protected]>
To: "Phil Neumiller" <[email protected]>; "Joe Touch"
<[email protected]>
Cc: <[email protected]>; "Behcet Sarikaya" <[email protected]>;
"Scott Corson" <[email protected]>; "Eric Whitehill"
<[email protected]>; <[email protected]>;
<[email protected]>; "Rex Buddenberg" <[email protected]>;
<[email protected]>; "Dr. Chane L. Fullmer" <[email protected]>;
<[email protected]>; "'William Ivancic'" <[email protected]>;
<[email protected]>; <[email protected]>; <[email protected]>;
<[email protected]>; "Anders Lindgren" <[email protected]>;
<[email protected]>; <[email protected]>; "Armando L. Caro Jr."
<[email protected]>; <[email protected]>
Sent: Thursday, June 06, 2002 1:50 PM
Subject: Re: [pilc] RE: AD request / L2 Triggers Chapter Statement


> Folks,
>   Thanks Joe for starting a good discussion on the
> Charter (not Chapter) statement that I had posted.
>   I had left the chairs open and so far no volunteers
> appeared to lead this activity. I approached some
> people but they did not seem to be interested in.
>   If anyone interested in leading please come forward
> so that some action can be put forward for IETF 54 or
> IETF 55.
>   Regards,
> --- Phil Neumiller <[email protected]>
> wrote:
> > > > Right, but many routing systems are built ON IP
> > and below transport.
> > > > Take MIP, any MANET protocol, ICMP itself, OSPF,
> > BGP4, etc.  All of
> > > > these systems that reside at the IP layer in
> > routers CAN INDEED make
> > > > extremely good use of this information.
> > >
> > > Which of these operates on millisecond decisions?
> >
> > Oh, this is a trick question....  NONE because they
> > CAN'T.  YOU can NOT
> > convince me that they WOULD NOT if they COULD!!!!
> >
> >
> > >
> > > What I'm asserting is:
> > >
> > > if it's within two links of the same type, a
> > meta-link can do it
> >
> > Yeah, and where is this defined in ANY commercial IP
> > stack???
> >
> >
> > > if it's between two different links, IP routing
> > has to do it,
> > > and that doesn't care about millisecond events
> >
> > Baloney!  Policy based routing most certainly does.
> > Especially with respect to
> > wireless handovers which is the MAIN purpose of this
> > discussion.  When users
> > are paying for mobile services they expect SERVICE.
> >
> > > TCP reacts on RTTs, usually multiple RTTs. Even
> > then, what would TCP do
> > > that it doesn't already to? I.e., you don't need
> > an L2 trigger to
> > > program a box to send TCP ECN data back.
> >
> > You are missing the point.  Most of us are talking
> > about wireless systems and the
> > L2 being able to convey its highly variable state to
> > stacks that CAN react to
> > it.
> >
> > > Again, if you want to do this on a single box, you
> > don't need an L2
> > > trigger. It's not that you're talking to IP in
> > general, but that you're
> > > talking to a particular stack implementation.
> >
> > Wrong.  We absolutely NEED IP to change so that
> > there is a standard
> > mechanism for ALL future L2s to communicate with IP
> > from the bottom up.
> > Imagine yourself as an L2.  How on earth are you
> > going to send any kind
> > of link quality assesment to IP?  THERE IS NO
> > STANDARD MECHANISM
> > CURRENTLY.  That is the reason we need the BOF, WG,
> > and the standard
> > track RFC for specifying the underspecified IP
> > underbelly.  The PILC WG has
> > taken an Ivory tower position and said all kinds of
> > things about what L2s
> > should and should not do.  I am representing L2s and
> > pushing back on what
> > I believe IP needs to do!!!  Its silly for the PILC
> > group to come up with all kinds
> > of recommendations for standards they can't control
> > when they CAN control
> > the safe harber that IP can provide for wireless and
> > other L2s.
> >
> > -Phil Neumiller
> >
> >
> >
> --behcet
>
> __________________________________________________
> Do You Yahoo!?
> Yahoo! - Official partner of 2002 FIFA World Cup
> http://fifaworldcup.yahoo.com
>

Suggested BoF Name: L2TG (L2 TriGgers)

BoF Description
---------------

Since the PILC working group was formed, an additional
set of problems involving the interaction between wireless
link layers and IP has emerged. These problems all have in
common the need for timely information from the link layer
about certain processes that are unique to wireless networks,
and which critically impact the performance or ease of implementation
at the IP layer. Such information has generically been called L2 triggers.
Specific examples of L2 triggers are information required to 
improve handover performance, particularly for real time traffic,
and to allow IP dormant mode host alerting (a.k.a paging)
to utilize link layer paging support. Drafts discussing these
particular problems are draft-manyfolks-mobilereq-l2-04.txt,
for handover, and draft-guri-seamoby-lahap-00.txt, for paging.
Additional possible uses of L2 triggers that have been
suggested are for MANETS and for radio QoS signaling. This 
BOF will discuss currently proposed and suggested uses for L2 triggers,
and whether enough work in this area exists to form a working
group.

Possible Topics for a Working Group Charter (should one be deemed desirable)
----------------------------------------------------------------

- Accept draft-manyfolks-mobilereq-l2.txt as a WG draft and 
complete to Informational RFC status.

- Accept that part of draft-guri-seamoby-lahap-00.txt dealing
with L2 triggers for paging and complete to Informational
RFC status.

- Solicit Internet drafts on L2 triggers of MANETs, in conjunction with
the MANET group, and radio QoS signaling, in conjunction with the NSIS group,
for completion to Informational RFC status.

- Develop and standardize an IP protocol for handover L2 triggers between
bridging access points and access routers. This would be for L2 protocols
in which a wired IP access network separates the access point from the 
access router.

- Discuss with IEEE 802.11, and any other SDO that might need it, the
possibility of including L2 triggers in their L2 protocols.