Bill, On Jun 19, 2008, at 8:01 PM, William Herrin wrote:
On Thu, Jun 19, 2008 at 10:30 PM, David Conrad <drc@virtualized.org> wrote:On Jun 19, 2008, at 10:48 PM, William Herrin wrote:1. The IPv6 variant (which may be but probably isn't compatible with IPv4) 2. The IPv4 variant (which is compatible with but likely suboptimal forIPv6) 3. The clean-slate variant which we could design if we weren't constrained by IPv4 or IPv6.In a world of limited resources, which should be the priority?The one that demonstrably impacts the largest installed customer base, obviously.
This isn't obvious to me.In the ideal world, we could come up with a solution that treats the end point identifier as a variable length opaque bag of bits and Do the Right Thing with it.
However, assuming we're not in the ideal world, there has to be some prioritization. The installed base is using a protocol that cannot meet the fundamental requirement of the Internet, that of providing ubiquitous connectivity. IPv4 augmented with NAT may be able to meet that requirement, but at some additional operational cost that many believe to be prohibitive. As such, it one must choose, it would seem obvious to me that you choose to modify the protocol that has the best chance of meeting long term requirements.
However, with that said, the basic question was whether or not there was (rough) consensus. It would appear not.
Regards, -drc -- to unsubscribe send a message to rrg-request@psg.com with the word 'unsubscribe' in a single line as the message text body. archive: <http://psg.com/lists/rrg/> & ftp://psg.com/pub/lists/rrg