RE: Off topic: Any one know of a good IPV6 reference book?

[Dave Piscitello <dave () corecom com>]

Ten Years After...

I was an unfortunate participant in IPv6 standards development. I didn't
support v6 then and have no more faith now (truth be told, I wouldn't support
what I supported then, now, either!)

v6 doesn't restore end-to-end computing. Maybe it solves a globally unique 
addressing issue, which is important to the mobile communications folks, 
but that's a different beast.

I honestly don't see how or more importantly why we'd stop using proxies, NAT
(esp the many-to-one address mappings at firewalls to hide internal addresses).
IPv6 deployment has been so lethargic we've added just about everything to 
IPv4.

Frankly, Marcus' assessment is spot on. And I recall there was an 
internet-draft long long time ago that suggested much of the same minor 
tweaks Marcus identifies, I don't know if it's anywhere to be found but 
Paul Frances, Steve Bellovin, or Steve Deering might know.

It was lightly regarded as and dismissed as it did not push the envelope 
far enough: it only fixed known problems and did not attempt to expand the role
and utility of IP into the next generation (historical note: all the IPv6 
candidates
were considered under the IPng working group).

I think Marcus's "what if they gave a protocol and nobody came?" is 
slightly off mark. It is likely to be a very lame party with an exhorbitant 
admission fee: bad food and drink, and too little of it.

At 07:04 PM 7/27/2003 -0400, Marcus J Ranum wrote:
> Irwin Lazar wrote:
> >IPv6 would allow for a restoration of true end-to-end computing instead 
> of requiring complex gateways, proxies, and NAT devices
>
> IPV6 is insane overcomplexity if that was the only problems we
> wanted to solve. We could have doubled the address size of V4,
> bumped the version number, and left-filled from zero. As far as the
> "route glut" problems that stimulated the original design of IPV6,
> we could have used conventions (e.g. something like CIDR addressing
> which hadn't been thought of when the V6 effort started) that could
> easily have solved those issues.
>
> Basically, the standards pukes are having fun playing their little
> games but none of it's really going to solve real problems. IPV6
> is gonna be like ISO protocols all over again: what if they gave
> a protocol and nobody came?
>
> mjr.
>
> _______________________________________________
> firewall-wizards mailing list
> firewall-wizards () honor icsalabs com
> http://honor.icsalabs.com/mailman/listinfo/firewall-wizards


----------
David M. Piscitello
Core Competence, Inc.
Myrtle Bank Lane HHI, SC 29926
Company: http://www.corecom.com
WebLog: http://hhi.corecom.com/weblogindex.htm
Personal: http://hhi.corecom.com



_______________________________________________
firewall-wizards mailing list
firewall-wizards () honor icsalabs com
http://honor.icsalabs.com/mailman/listinfo/firewall-wizards