Ladies and Gentlemen, First, please excuse my lack of background and intellectual knowledge in all this discussion.
Many of the comments that I have read over the past few weeks regarding this seem to revolve around the "theory" of how it should work. Theory is great. Many of the people in this group that post are from acedemia and research areas. I don't see too many posts from people that are actually trying to make this all work. The comments that "NAT shouldn't be used in IPv6 since we will have more than enough IPs" is also great, in THEORY! Did we all think that we would have enough IP numbers when IPv4 was started? I work for a federal agency that has over 6,000,000 devices that need IP numbers. Most need access to the "outside" world. However, do I want all of these devices visible to the out side world? NO! Yes, we have border routers and firewalls that block access from many of the "undesirables" that are out there. Even if we get enough IPv6 numbers to ensure that every device can have a unique number, we will still use NAT. Ok, federal government and leading edge technology go together like military intelligence. The use of site-local is a great proposal. What I think should be done is just like in the IPv4 world, reserve a block (or a couple of blocks) of numbers that are non-routable. This will allow companies to know what nubers they are to use when setting up site-local numbers based on the number of devices. Now, if I have misunderstood the entire context of this post, please refer back to my opening sentence. Gary Allmond -------------------------------------------------------------------- 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] --------------------------------------------------------------------
