BCP for multisite multihoming

[bmanning at karoshi.com]

On Mon, May 21, 2007 at 01:23:06PM -0700, David Conrad wrote:
> Bill,
> 
> On May 21, 2007, at 12:57 PM, bmanning at karoshi.com wrote:
> >
> >which problem are we "solving" here?
> 
> Well, from the subject of the message, I think it's "multisite  
> multihoming".

	ah... i was distracted by the shiny token of vast tracts of 
	wasted address space as a "simple" way to multihome... based
	on the unstated assumption that a large enough sized block 
	would not be filtered.  so sorry for the divergent vector.
	we now return you to burning up the address space faster than
	at any time in the past...

> What is relevant, at least to the question asked, is under existing  
> policies, your alternative proposal is?

	under existant policies - no other alternative can exist.
	-IF- RIR's were to emphatically re-assert that they can NOT
	assure routeablity and they were changing policy to delegate
	only the block size justified, it would kick out the plausable
	denyability crutch that ISPs use...  'Its the RIR's fault for
	passing out prefixes we can't stuff into a DF routing table'
	ISPs ought to make an informed and delibeate choice as to 
	what prefixes to carry...  the myth of the "global routing table"
	is a false and is a shaky basis on which to base a telecommunications
	business.
	
	But i will take off my rose-coloured glasses, step off my soap box
	and suggest the more pragmatic approch.

	as stated earlier... ramp up a design team as soon as possible
	to design the successor to IPv6... its not going to last nearly
	as long as IPv4.   

> 
> Rgds,
> -drc