On Tue, Feb 01, 2005 at 05:39:15PM +0200, Pekka Savola wrote: > On Tue, 1 Feb 2005, Bob Hinden wrote: > >My take of this is that they should remain in the IPv6 address > >architecture. There is current usage and removing them would break other > >specifications. > > I would agree with that conclusion for mapped addresses, but I have > heard NO ONE explicitly saying anything about the usefulness of > compatible addresses. > > Thus my take is that compatibles should be removed, and some kind of > warning/reference text added to the mapped addresses. > draft-ietf-v6ops-application-transition-03 (soon to be RFC) discusses > some of this.
Hi Pekka, I thought compatibles had (or were) being removed. That's why all reference to them was removed from the new transition mechanisms RFC update? See section 9 of draft-ietf-v6ops-mech-v2-06.txt. If we're doing a u-turn on that, we should catch it in this draft too. The docs at http://gsyc.escet.urjc.es/~eva/IPv6-web/ipv6.html and as draft-ietf-v6ops-application-transition-03.txt (not -02!) show good practice. Has the application-transition draft been last called yet? Could we get it pushed and cited here in this RFC update? I didn't realise mapped was disabled on so many systems, very interesting. I am indifferent on mapped addresses; they clearly have use, and are being used, but it's not a wholly 'clean' solution for some of the reasons posted here and were we to start again... I think Itojun's mapped-on-the-wire-harmful draft is also good to cite, but that seems unlikely to ever complete? -- Tim -------------------------------------------------------------------- IETF IPv6 working group mailing list [email protected] Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6 --------------------------------------------------------------------
