Re: W.G. Last Call on "IP Version 6 Addressing Architecture"

[Randy Bush]

> So "unassigned" means really "assigned for unicast, but undefined yet
> type of addresses".
> 
> That would certainly not represent accurately or encourage any future
> developments other than the one focused on unicast. How could we know
> now what the future communication models would be? Furthermore, I do not
> see the motivation behind such restriction.
> 
> Could you elaborate on the exact reasons?

because we remember implementations and protocols that had A, B, and C hard
coded and how hard it was getting cidr deployed.

we want to make it very clear that current/new implementations should treat
it all as unicast.

if someone comes up with some bright idea down the pike, then is the time to
hack.  but we should not set things up so we have to hack to achieve what we
see as the most probably path today.

make sense?

randy
--------------------------------------------------------------------
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]
--------------------------------------------------------------------