poty@iiat.ru wrote:
Maybe you are right, but it doesn’t prove that is IS good in IPv6 world too. I can’t understand, why I should think about such private matters (and indirectly fund this) and count it as arguments in the RIPE’s policy development? If the allocation will never be announced to the public network called the Internet, then it’s not the scope of our thinking!
IPv4: * RFC1918 + just grab - everybody in the world uses it, lots of clashes - not suitable for interconnecting ever to other networks - generally implies a lot of NAT at one point in time * RIR-space + guaranteed globally unique - you will have to do paperwork and pay for it IPv6: * ULA, RFC4193 + nobody to talk to, calculate your own - never to be used anywhere on the Internet - not 100.00000% sure that it is globally unique (also see http://www.sixxs.net/tools/grh/ula/ for a 'registry' which would make it at least "unique" when everybody uses that) - could imply NAT, though that should not be used with IPv6 * RIR-space + guaranteed globally unique + can be routed on the internet - you will have to do paperwork and pay for it You can pick what you want, but heed the warnings. Greets, Jeroen