Re: Why, technically, MIP and IPv6 can't be deployed

I think that this begs the question of where the
larger problem lies: while IP can run over pigeons,
bailing wire and quite possibly chewing gum, there
are clearly some media that IP runs over better. Looking
at the various wireless media, they are either somewhere
between completely hopeless (cellular with its purpose
built PSTN roots) to "overhelpful" (bluetooth) to sort
of ok for a first order approximation (802.11). IMO, what we
need are radio networks whose L1/L2's are built with IP in
mind first and foremost -- not afterthoughts, not "one of
many". 

There are surely some things that IP -- both v4 and v6 -- could do
better, but what they will never be able to do is 
work well over a broken L1/L2. Until we have  wireless
media which are trying to work with rather than at cross
purposes with IP, it really doesn't matter what IETF 
specifies.

Finally: I rather get annoyed when L1/L2 people tell me
"that's not the way our L1/L2 works!", blah, blah, blah.
Fine. Engineer us something that does work; stop telling
us to engineer for broken media. IP has won in the
marketplace in case they have been asleep for the last 10 years.

                Mike

On Tue, 2004-11-09 at 04:14, Masataka Ohta wrote:
> Tim Chown wrote:
>
> IPv6 is defective in so many ways. But, w.r.t. WLAN, here is the
> reason.
>
> > Could you describe why exactly IPv6 can't run on the (layer 2?) WLAN
> > infrastructure?
> That ND extensively, without any valid reason to do so, use
> multicast, which is not acknowledged at WLAN L2, means IPv6
> or its ND is unreliable over congested WLAN. If multicast
> ND packet is lost by congestion, it is not retransmitted by L2.
>
> MIP failed mostly because there has been no standards MIP over
> link technologies, which are adaptation layers between L2 and L3.
>
> RFC2002 does not specify anything about how CoA addresses can
> be obtained, which is fine, if and only if there are other
> specifications on link or provider dependent ways to do so.
>
> IPv6 made it worse by trying to standardize ND as *THE* adaptation
> mechanism, even though the mechanism *MUST* depend on details of
> L2 and *MUST* be different L2 by L2.
>
> As a result, IPv6 won't stably run over WLAN, multicast over which
> has characteristics much different from wired Ether.
>
>                                               Masataka Ohta
>
>
>
> _______________________________________________
> Ietf mailing list
> Ietf(_at_)ietf(_dot_)org
> https://www1.ietf.org/mailman/listinfo/ietf

_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/ietf