On Fri, 2009-12-11 at 21:45 -0800, Roger Marquis wrote:
If you're going to implement statefulness there is no technical downside to implementing NAT as well. No downside, plenty of upsides, no brainer...
Of course there are downsides to implementing NAT - adding any feature to a device increases its complexity and affects its expense, time to market, MTBF etc. And there is certainly a downside to *deploying* NAT: NAT removes end-to-end transparency. Gotta keep those SOHO users in their cages, don't want them becoming independent producers of digital value, no sir! Seriously - by all means keep NAT as a technology for those who want to deploy it; we can't uninvent it anyway. It just shouldn't be imposed on others. I would argue that an ISP requiring of a customer that they use a NATted solution with IPv6 *is* imposing it on others. Regards, K. -- ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Karl Auer (kauer@biplane.com.au) +61-2-64957160 (h) http://www.biplane.com.au/~kauer/ +61-428-957160 (mob) GPG fingerprint: 07F3 1DF9 9D45 8BCD 7DD5 00CE 4A44 6A03 F43A 7DEF