IMHO, there are two ways to waste address space. One worse (again, IMHO) than the other. One mechanism is to distribute it so liberally that it causes a detriment to those who need it later. This is what is being argued by Dean and others. However, I don’t think that’s possible with what I propose, since I suggest that if we burn through 1/8th of the total address space using such liberal policies in less than 50 years, we should, in fact, consider more conservative policies for the remaining 3/4. (another 1/8th is partially consumed by things like multicast, link local, ULA, etc.). A far worse mechanism to waste address space is to create hardships now for organizations with actual present needs or unnecessary human factors errors, etc. just for the sake of keeping a large supply of addresses on the shelf until some point well past the useful life of the protocol. I find it very unlikely that IPv6 will remain the protocol of choice for the lingua franca of the internet for much more than 100 years. (Frankly, I think 50 is probably a stretch, but that’s about how long it looks like IPv4 will end up lasting). I think we will hit many other limitations of IPv6 well before we run out of addresses even with the level of liberalized allocation policy that I have supported and continue to support. Owen On Sep 17, 2014, at 7:02 PM, Skeeve Stevens <skeeve at v4now dot com> wrote:
|