> > > > > I'm not sure what we gain by doing that, as opposed to setting aside
> > > > > private address space from any global prefix by filtering it at
> > > > > administrative boundaries...
> > > >
> > > > we need the ability to assign global prefixes to sites that aren't
> > > > directly connected to the public Internet, even though they might
> > > > be connected to large numbers of other networks.
> > >
> > > Allocating one /32 and giving these sites a /48 each should last for the
> > > first 64k such private networks (which are assumed to be rather rare).
> >
> > they might be more plentiful than you think. [...]
>
> Well, personally, I can clearly see a few cases (e.g. in the military),
> but the rest, I could separate to a few cases:
>
> 1) have IPv4 connectivity but no IPv6 (or stable IPv6 -- something like
> 6to4 with dynamic v4 address)
>
> 2) have zero 'net connectivity ("clear cases") -- usually by design
>
> 3) have full connectivity but want somehow permanent addresses, to be used
> locally, for renumbering, to be leaked to the internet, etc.etc.
4) have zero connectivity to the public internet (by choice) but have
interconnections with an arbitrary number of other private networks
(which might or might not have connectivity to the public internet).
this is actually fairly common. we need to get away from the assumption
that the reason that people use IP is to connect to the public Internet.
Keith
--------------------------------------------------------------------
IETF IPng Working Group Mailing List
IPng Home Page: http://playground.sun.com/ipng
FTP archive: ftp://playground.sun.com/pub/ipng
Direct all administrative requests to [EMAIL PROTECTED]
--------------------------------------------------------------------
