On Thu Aug 23 21:12:17 2007, Sam Hartman wrote:
>>>>> "Keith" == Keith Moore <moore(_at_)cs(_dot_)utk(_dot_)edu> writes:
Keith> Hallam-Baker, Phillip wrote:
>> If we can meet the needs of 80% of Internet users with some
>> form of shared access there will be more addresses left for
>> 20% with greater needs.
>> Keith> with 2**128 potential addresses, this is not only
Keith> unnecessary, it's harmful. there's far greater benefit
Keith> be had by uniformity in address allocation, globally
Keith> addresses, and consistent use of addresses end-to-end.
I'll take ease in renumbering over application transparency for any
I find this confusing as a concern - how often do you renumber?
Compare this to how often you run applications which benefit from
transparency. I'm pretty sure I've run FTP, for example, at least
twice since I last renumbered. I've even run VOIP - another prime
consumer of network transparency.
I agree that renumbering isn't a trivial exercise, but if your
external address changes in NAT, it's just as painful (if not
slightly more so) for externally facing services, surely? And with
IPv6, doesn't renumbering get easier? (I thought it did, when I
renumbered mine, and I had to do so at very short notice).
I'm quite possibly being naïve, or stupid, or all three, but I don't
follow why renumbering is a sufficiently major concern that
applications must suffer.
Dave Cridland - mailto:dave(_at_)cridland(_dot_)net -
Infotrope Polymer - ACAP, IMAP, ESMTP, and Lemonade
Ietf mailing list